#373 by GitHub stars
github/squad
Squad: AI agent teams for any project
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
.copilot/skills/action-skill-contracts/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/agent-collaboration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/agent-conduct/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/architectural-proposals/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/architectural-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/archival-integrity/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/ci-validation-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/cli-wiring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/client-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/content-scope/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/distributed-mesh/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/documentation-maintenance/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/git-workflow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/github-multi-account/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/history-hygiene/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/pr-lifecycle/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/protected-files/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/release-process/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/reskill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/reviewer-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/secret-handling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/security-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/squad-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/test-discipline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.copilot/skills/windows-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/agent-collaboration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/agent-conduct/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/architectural-proposals/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/ci-validation-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/cli-wiring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/client-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/coordinator-init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/coordinator-response-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/coordinator-source-of-truth/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/cross-machine-coordination/SKILL.md
Week of Nov 2, 2026 · +35 / −16 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
- 3 description: "Enables squad agents on different machines to share work via git-based task queuing"
+ 3 description: "Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002)."
- 5 confidence: "medium"
+ 5 confidence: "low"
- 14 **Status:** Specification (ready for implementation)
+ 14 **Status:** Design proposal — NOT implemented
+ 15
+ 16 > ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
+ 17 > the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
+ 18 > and `squad watch` do not poll task files, run whitelisted commands, or
+ 19 > write result files. Everything below describes a proposed design, not
+ 20 > current behavior. Do not tell users it works, and do not create task files
+ 21 > expecting another machine to pick them up.
+ 22 >
+ 23 > **Use these instead:**
+ 24 >
+ 25 > - **Several machines or pods running `squad watch --execute` on one
+ 26 > repository:** enable work claims (`squad watch --execute --claims
+ 27 > git-ref`, or `watch.claims` in `.squad/config.json`). Claims are Git-ref
+ 28 > leases that make sure only one watcher works on each issue. See ADR-002
+ 29 > (`docs/specification/adr-002-work-claims-and-external-queues.md`) and the
+ 30 > "Multi-Machine Watch with Work Claims" docs scenario
+ 31 > (`docs/src/content/docs/scenarios/multi-machine-watch.md`).
+ 32 > - **Sending work to a machine with specific hardware:** use capability
+ 33 > routing with `needs:*` issue labels and a machine capabilities file.
- 56 Ralph on DevBox will:
+ 75 In this proposal (not implemented), Ralph on DevBox would:
- 67 Ralph automatically watches `.squad/cross-machine/tasks/` for work targeted at this machine.
- 68
- 69 **On each cycle (5-10 min):**
+ 86 In this proposal (not implemented), Ralph would watch `.squad/cross-machine/tasks/` for work targeted at this machine. Ralph does not do this today.
+ 87
+ 88 **On each cycle (5-10 min), as proposed:**
- 72 # Pseudo-code (Ralph implementation)
+ 91 # Pseudo-code (proposed Ralph implementation; does not exist)
- 99 Ralph on DevBox will:
+ 118 In this proposal (not implemented), Ralph on DevBox would:
- 207 Ralph reads config from `.squad/config.json`:
+ 226 Proposed `.squad/config.json` block. No Squad release reads it:
- 390 Ralph automatically includes this pattern in its watch loop:
- 391
- 392 ```
- 393 Ralph Watch Cycle (every 5-10 min):
+ 409 Not implemented: `squad watch` does not include this pattern. The proposed watch loop was:
+ 410
+ 411 ```
+ 412 Proposed Ralph Watch Cycle (every 5-10 min):
- 405 No manual Ralph configuration needed — just create task files or issues with the right labels.
+ 424 Until this is built, task files and `squad:machine-*` labels do nothing. To stop several watchers from picking the same issue, use work claims (`--claims git-ref`, ADR-002).
- 417 **After (with this pattern):**
+ 436 **After (if this proposal is built):**
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/cross-squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/cross-squad-communication/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/distributed-mesh/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/docs-standards/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/e2e-template-testing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/economy-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/error-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/external-comms/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/fact-checking/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/gh-auth-isolation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/gh-aw-enlistment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/git-workflow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/github-multi-account/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/history-hygiene/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/humanizer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/iterative-retrieval/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/nap/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/notification-routing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/pr-review-response/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/pr-screenshots/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/project-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/ralph-two-pass-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/reflect/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/release-process/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/reskill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/retro-enforcement/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/reviewer-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/secret-handling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/session-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/squad-commands/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/squad-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/squad-help/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/squad-version-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/test-discipline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/tiered-memory/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/versioning-policy/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad-templates/skills/windows-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad/skills/team-memory-maintenance/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad/skills/work-monitoring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
.squad/templates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/agent-collaboration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/agent-conduct/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/architectural-proposals/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/ci-validation-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/cli-wiring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/client-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/coordinator-init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/coordinator-response-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/coordinator-source-of-truth/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/cross-machine-coordination/SKILL.md
Week of Nov 2, 2026 · +35 / −16 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
- 3 description: "Enables squad agents on different machines to share work via git-based task queuing"
+ 3 description: "Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002)."
- 5 confidence: "medium"
+ 5 confidence: "low"
- 14 **Status:** Specification (ready for implementation)
+ 14 **Status:** Design proposal — NOT implemented
+ 15
+ 16 > ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
+ 17 > the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
+ 18 > and `squad watch` do not poll task files, run whitelisted commands, or
+ 19 > write result files. Everything below describes a proposed design, not
+ 20 > current behavior. Do not tell users it works, and do not create task files
+ 21 > expecting another machine to pick them up.
+ 22 >
+ 23 > **Use these instead:**
+ 24 >
+ 25 > - **Several machines or pods running `squad watch --execute` on one
+ 26 > repository:** enable work claims (`squad watch --execute --claims
+ 27 > git-ref`, or `watch.claims` in `.squad/config.json`). Claims are Git-ref
+ 28 > leases that make sure only one watcher works on each issue. See ADR-002
+ 29 > (`docs/specification/adr-002-work-claims-and-external-queues.md`) and the
+ 30 > "Multi-Machine Watch with Work Claims" docs scenario
+ 31 > (`docs/src/content/docs/scenarios/multi-machine-watch.md`).
+ 32 > - **Sending work to a machine with specific hardware:** use capability
+ 33 > routing with `needs:*` issue labels and a machine capabilities file.
- 56 Ralph on DevBox will:
+ 75 In this proposal (not implemented), Ralph on DevBox would:
- 67 Ralph automatically watches `.squad/cross-machine/tasks/` for work targeted at this machine.
- 68
- 69 **On each cycle (5-10 min):**
+ 86 In this proposal (not implemented), Ralph would watch `.squad/cross-machine/tasks/` for work targeted at this machine. Ralph does not do this today.
+ 87
+ 88 **On each cycle (5-10 min), as proposed:**
- 72 # Pseudo-code (Ralph implementation)
+ 91 # Pseudo-code (proposed Ralph implementation; does not exist)
- 99 Ralph on DevBox will:
+ 118 In this proposal (not implemented), Ralph on DevBox would:
- 207 Ralph reads config from `.squad/config.json`:
+ 226 Proposed `.squad/config.json` block. No Squad release reads it:
- 390 Ralph automatically includes this pattern in its watch loop:
- 391
- 392 ```
- 393 Ralph Watch Cycle (every 5-10 min):
+ 409 Not implemented: `squad watch` does not include this pattern. The proposed watch loop was:
+ 410
+ 411 ```
+ 412 Proposed Ralph Watch Cycle (every 5-10 min):
- 405 No manual Ralph configuration needed — just create task files or issues with the right labels.
+ 424 Until this is built, task files and `squad:machine-*` labels do nothing. To stop several watchers from picking the same issue, use work claims (`--claims git-ref`, ADR-002).
- 417 **After (with this pattern):**
+ 436 **After (if this proposal is built):**
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/cross-squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/cross-squad-communication/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/distributed-mesh/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/docs-standards/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/e2e-template-testing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/economy-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/error-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/external-comms/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/fact-checking/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/gh-auth-isolation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/gh-aw-enlistment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/git-workflow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/github-multi-account/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/history-hygiene/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/humanizer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/iterative-retrieval/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/nap/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/notification-routing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/pr-review-response/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/pr-screenshots/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/project-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/ralph-two-pass-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/reflect/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/release-process/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/reskill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/retro-enforcement/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/reviewer-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/secret-handling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/session-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/squad-commands/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/squad-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/squad-help/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/squad-version-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/test-discipline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/tiered-memory/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/versioning-policy/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-cli/templates/skills/windows-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/agent-collaboration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/agent-conduct/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/architectural-proposals/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/ci-validation-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/cli-wiring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/client-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/coordinator-init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/coordinator-response-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/coordinator-source-of-truth/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/cross-machine-coordination/SKILL.md
Week of Nov 2, 2026 · +35 / −16 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
- 3 description: "Enables squad agents on different machines to share work via git-based task queuing"
+ 3 description: "Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002)."
- 5 confidence: "medium"
+ 5 confidence: "low"
- 14 **Status:** Specification (ready for implementation)
+ 14 **Status:** Design proposal — NOT implemented
+ 15
+ 16 > ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
+ 17 > the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
+ 18 > and `squad watch` do not poll task files, run whitelisted commands, or
+ 19 > write result files. Everything below describes a proposed design, not
+ 20 > current behavior. Do not tell users it works, and do not create task files
+ 21 > expecting another machine to pick them up.
+ 22 >
+ 23 > **Use these instead:**
+ 24 >
+ 25 > - **Several machines or pods running `squad watch --execute` on one
+ 26 > repository:** enable work claims (`squad watch --execute --claims
+ 27 > git-ref`, or `watch.claims` in `.squad/config.json`). Claims are Git-ref
+ 28 > leases that make sure only one watcher works on each issue. See ADR-002
+ 29 > (`docs/specification/adr-002-work-claims-and-external-queues.md`) and the
+ 30 > "Multi-Machine Watch with Work Claims" docs scenario
+ 31 > (`docs/src/content/docs/scenarios/multi-machine-watch.md`).
+ 32 > - **Sending work to a machine with specific hardware:** use capability
+ 33 > routing with `needs:*` issue labels and a machine capabilities file.
- 56 Ralph on DevBox will:
+ 75 In this proposal (not implemented), Ralph on DevBox would:
- 67 Ralph automatically watches `.squad/cross-machine/tasks/` for work targeted at this machine.
- 68
- 69 **On each cycle (5-10 min):**
+ 86 In this proposal (not implemented), Ralph would watch `.squad/cross-machine/tasks/` for work targeted at this machine. Ralph does not do this today.
+ 87
+ 88 **On each cycle (5-10 min), as proposed:**
- 72 # Pseudo-code (Ralph implementation)
+ 91 # Pseudo-code (proposed Ralph implementation; does not exist)
- 99 Ralph on DevBox will:
+ 118 In this proposal (not implemented), Ralph on DevBox would:
- 207 Ralph reads config from `.squad/config.json`:
+ 226 Proposed `.squad/config.json` block. No Squad release reads it:
- 390 Ralph automatically includes this pattern in its watch loop:
- 391
- 392 ```
- 393 Ralph Watch Cycle (every 5-10 min):
+ 409 Not implemented: `squad watch` does not include this pattern. The proposed watch loop was:
+ 410
+ 411 ```
+ 412 Proposed Ralph Watch Cycle (every 5-10 min):
- 405 No manual Ralph configuration needed — just create task files or issues with the right labels.
+ 424 Until this is built, task files and `squad:machine-*` labels do nothing. To stop several watchers from picking the same issue, use work claims (`--claims git-ref`, ADR-002).
- 417 **After (with this pattern):**
+ 436 **After (if this proposal is built):**
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/cross-squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/cross-squad-communication/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/distributed-mesh/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/docs-standards/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/e2e-template-testing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/economy-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/error-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/external-comms/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/fact-checking/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/gh-auth-isolation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/gh-aw-enlistment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/git-workflow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/github-multi-account/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/history-hygiene/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/humanizer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/iterative-retrieval/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/nap/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/notification-routing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/pr-review-response/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/pr-screenshots/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/project-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/ralph-two-pass-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/reflect/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/release-process/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/reskill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/retro-enforcement/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/reviewer-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/secret-handling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/session-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/squad-commands/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/squad-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/squad-help/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/squad-version-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/test-discipline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/tiered-memory/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/versioning-policy/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
packages/squad-sdk/templates/skills/windows-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/agent-collaboration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/agent-conduct/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/architectural-proposals/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/ci-validation-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/cli-wiring/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/client-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/coordinator-init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/coordinator-response-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/coordinator-source-of-truth/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/cross-machine-coordination/SKILL.md
Week of Nov 2, 2026 · +35 / −16 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
- 3 description: "Enables squad agents on different machines to share work via git-based task queuing"
+ 3 description: "Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002)."
- 5 confidence: "medium"
+ 5 confidence: "low"
- 14 **Status:** Specification (ready for implementation)
+ 14 **Status:** Design proposal — NOT implemented
+ 15
+ 16 > ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
+ 17 > the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
+ 18 > and `squad watch` do not poll task files, run whitelisted commands, or
+ 19 > write result files. Everything below describes a proposed design, not
+ 20 > current behavior. Do not tell users it works, and do not create task files
+ 21 > expecting another machine to pick them up.
+ 22 >
+ 23 > **Use these instead:**
+ 24 >
+ 25 > - **Several machines or pods running `squad watch --execute` on one
+ 26 > repository:** enable work claims (`squad watch --execute --claims
+ 27 > git-ref`, or `watch.claims` in `.squad/config.json`). Claims are Git-ref
+ 28 > leases that make sure only one watcher works on each issue. See ADR-002
+ 29 > (`docs/specification/adr-002-work-claims-and-external-queues.md`) and the
+ 30 > "Multi-Machine Watch with Work Claims" docs scenario
+ 31 > (`docs/src/content/docs/scenarios/multi-machine-watch.md`).
+ 32 > - **Sending work to a machine with specific hardware:** use capability
+ 33 > routing with `needs:*` issue labels and a machine capabilities file.
- 56 Ralph on DevBox will:
+ 75 In this proposal (not implemented), Ralph on DevBox would:
- 67 Ralph automatically watches `.squad/cross-machine/tasks/` for work targeted at this machine.
- 68
- 69 **On each cycle (5-10 min):**
+ 86 In this proposal (not implemented), Ralph would watch `.squad/cross-machine/tasks/` for work targeted at this machine. Ralph does not do this today.
+ 87
+ 88 **On each cycle (5-10 min), as proposed:**
- 72 # Pseudo-code (Ralph implementation)
+ 91 # Pseudo-code (proposed Ralph implementation; does not exist)
- 99 Ralph on DevBox will:
+ 118 In this proposal (not implemented), Ralph on DevBox would:
- 207 Ralph reads config from `.squad/config.json`:
+ 226 Proposed `.squad/config.json` block. No Squad release reads it:
- 390 Ralph automatically includes this pattern in its watch loop:
- 391
- 392 ```
- 393 Ralph Watch Cycle (every 5-10 min):
+ 409 Not implemented: `squad watch` does not include this pattern. The proposed watch loop was:
+ 410
+ 411 ```
+ 412 Proposed Ralph Watch Cycle (every 5-10 min):
- 405 No manual Ralph configuration needed — just create task files or issues with the right labels.
+ 424 Until this is built, task files and `squad:machine-*` labels do nothing. To stop several watchers from picking the same issue, use work claims (`--claims git-ref`, ADR-002).
- 417 **After (with this pattern):**
+ 436 **After (if this proposal is built):**
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/cross-squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/cross-squad-communication/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/distributed-mesh/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/docs-standards/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/e2e-template-testing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/economy-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/error-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/external-comms/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/fact-checking/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/gh-auth-isolation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/gh-aw-enlistment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/git-workflow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/github-multi-account/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/history-hygiene/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/humanizer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/init-mode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/iterative-retrieval/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/model-selection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/nap/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/notification-routing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/pr-review-response/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/pr-screenshots/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/project-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/ralph-two-pass-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/reflect/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/release-process/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/reskill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/retro-enforcement/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/reviewer-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/secret-handling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/session-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/squad/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/squad-commands/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/squad-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/squad-help/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/squad-version-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/test-discipline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/tiered-memory/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/versioning-policy/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
templates/skills/windows-compatibility/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
test-fixtures/skills-test-abb30454/bad-skill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
test-fixtures/skills-test-abb30454/my-skill/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
workflows/skills/gh-aw-enlistment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Network or broad shell
2aa2cc41b4d8 → 12fbcb5ac4b4
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 2aa2cc41b4d8
No earlier version was observed for comparison.
View on GitHubAdd to my skills
Verified commit 12fbcb5ac4b4075abb5d2ce743f4ba1be77e0200
git clone --filter=blob:none --no-checkout https://github.com/github/squad.git .hotskills/github-squad && git -C .hotskills/github-squad checkout --detach 12fbcb5ac4b4075abb5d2ce743f4ba1be77e0200
# Copy each selected skill directory from .hotskills/github-squad into .github/skills/<skill-name>/3.3kGitHub stars
25stars in 7 days
238skills in this repository
141commits in 30 days
Heat
6.7- Star growth, last 7 daysAhead of 79% of tracked repositories
- Star growth, last 24 hoursAhead of 94% of tracked repositories
- Rank movement, last 7 daysAhead of 43% of tracked repositories
- Commits, last 30 daysAhead of 95% of tracked repositories
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekAhead of 0% of tracked repositories
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- action-skill-contracts
.copilot/skills/action-skill-contracts - agent-collaboration
.copilot/skills/agent-collaboration - agent-conduct
.copilot/skills/agent-conduct - architectural-proposals
.copilot/skills/architectural-proposals - architectural-review
.copilot/skills/architectural-review - archival-integrity
.copilot/skills/archival-integrity - ci-validation-gates
.copilot/skills/ci-validation-gates - cli-wiring
.copilot/skills/cli-wiring - client-compatibility
.copilot/skills/client-compatibility - content-scope
.copilot/skills/content-scope - distributed-mesh
.copilot/skills/distributed-mesh - documentation-maintenance
.copilot/skills/documentation-maintenance - git-workflow
.copilot/skills/git-workflow - github-multi-account
.copilot/skills/github-multi-account - history-hygiene
.copilot/skills/history-hygiene - init-mode
.copilot/skills/init-mode - model-selection
.copilot/skills/model-selection - pr-lifecycle
.copilot/skills/pr-lifecycle - protected-files
.copilot/skills/protected-files - release-process
.copilot/skills/release-process - reskill
.copilot/skills/reskill - reviewer-protocol
.copilot/skills/reviewer-protocol - secret-handling
.copilot/skills/secret-handling - security-review
.copilot/skills/security-review - squad-conventions
.copilot/skills/squad-conventions - test-discipline
.copilot/skills/test-discipline - windows-compatibility
.copilot/skills/windows-compatibility - {skill-name}
.squad-templates - agent-collaboration
.squad-templates/skills/agent-collaboration - agent-conduct
.squad-templates/skills/agent-conduct - architectural-proposals
.squad-templates/skills/architectural-proposals - ci-validation-gates
.squad-templates/skills/ci-validation-gates - cli-wiring
.squad-templates/skills/cli-wiring - client-compatibility
.squad-templates/skills/client-compatibility - coordinator-init-mode
.squad-templates/skills/coordinator-init-mode - coordinator-response-mode
.squad-templates/skills/coordinator-response-mode - coordinator-source-of-truth
.squad-templates/skills/coordinator-source-of-truth - cross-machine-coordination
.squad-templates/skills/cross-machine-coordination - cross-squad
.squad-templates/skills/cross-squad - cross-squad-communication
.squad-templates/skills/cross-squad-communication - distributed-mesh
.squad-templates/skills/distributed-mesh - docs-standards
.squad-templates/skills/docs-standards - e2e-template-testing
.squad-templates/skills/e2e-template-testing - economy-mode
.squad-templates/skills/economy-mode - error-recovery
.squad-templates/skills/error-recovery - external-comms
.squad-templates/skills/external-comms - fact-checking
.squad-templates/skills/fact-checking - gh-auth-isolation
.squad-templates/skills/gh-auth-isolation - gh-aw-enlistment
.squad-templates/skills/gh-aw-enlistment - git-workflow
.squad-templates/skills/git-workflow - github-multi-account
.squad-templates/skills/github-multi-account - history-hygiene
.squad-templates/skills/history-hygiene - humanizer
.squad-templates/skills/humanizer - init-mode
.squad-templates/skills/init-mode - iterative-retrieval
.squad-templates/skills/iterative-retrieval - model-selection
.squad-templates/skills/model-selection - nap
.squad-templates/skills/nap - notification-routing
.squad-templates/skills/notification-routing - pr-review-response
.squad-templates/skills/pr-review-response - pr-screenshots
.squad-templates/skills/pr-screenshots - project-conventions
.squad-templates/skills/project-conventions - ralph-two-pass-scan
.squad-templates/skills/ralph-two-pass-scan - reflect
.squad-templates/skills/reflect - release-process
.squad-templates/skills/release-process - reskill
.squad-templates/skills/reskill - retro-enforcement
.squad-templates/skills/retro-enforcement - reviewer-protocol
.squad-templates/skills/reviewer-protocol - secret-handling
.squad-templates/skills/secret-handling - session-recovery
.squad-templates/skills/session-recovery - squad
.squad-templates/skills/squad - squad-commands
.squad-templates/skills/squad-commands - squad-conventions
.squad-templates/skills/squad-conventions - squad-help
.squad-templates/skills/squad-help - squad-version-check
.squad-templates/skills/squad-version-check - test-discipline
.squad-templates/skills/test-discipline - tiered-memory
.squad-templates/skills/tiered-memory - versioning-policy
.squad-templates/skills/versioning-policy - windows-compatibility
.squad-templates/skills/windows-compatibility - model-selection
.squad/skills/model-selection - team-memory-maintenance
.squad/skills/team-memory-maintenance - work-monitoring
.squad/skills/work-monitoring - {skill-name}
.squad/templates - {skill-name}
packages/squad-cli/templates - agent-collaboration
packages/squad-cli/templates/skills/agent-collaboration - agent-conduct
packages/squad-cli/templates/skills/agent-conduct - architectural-proposals
packages/squad-cli/templates/skills/architectural-proposals - ci-validation-gates
packages/squad-cli/templates/skills/ci-validation-gates - cli-wiring
packages/squad-cli/templates/skills/cli-wiring - client-compatibility
packages/squad-cli/templates/skills/client-compatibility - coordinator-init-mode
packages/squad-cli/templates/skills/coordinator-init-mode - coordinator-response-mode
packages/squad-cli/templates/skills/coordinator-response-mode - coordinator-source-of-truth
packages/squad-cli/templates/skills/coordinator-source-of-truth - cross-machine-coordination
packages/squad-cli/templates/skills/cross-machine-coordination - cross-squad
packages/squad-cli/templates/skills/cross-squad - cross-squad-communication
packages/squad-cli/templates/skills/cross-squad-communication - distributed-mesh
packages/squad-cli/templates/skills/distributed-mesh - docs-standards
packages/squad-cli/templates/skills/docs-standards - e2e-template-testing
packages/squad-cli/templates/skills/e2e-template-testing - economy-mode
packages/squad-cli/templates/skills/economy-mode - error-recovery
packages/squad-cli/templates/skills/error-recovery - external-comms
packages/squad-cli/templates/skills/external-comms - fact-checking
packages/squad-cli/templates/skills/fact-checking - gh-auth-isolation
packages/squad-cli/templates/skills/gh-auth-isolation - gh-aw-enlistment
packages/squad-cli/templates/skills/gh-aw-enlistment - git-workflow
packages/squad-cli/templates/skills/git-workflow - github-multi-account
packages/squad-cli/templates/skills/github-multi-account - history-hygiene
packages/squad-cli/templates/skills/history-hygiene - humanizer
packages/squad-cli/templates/skills/humanizer - init-mode
packages/squad-cli/templates/skills/init-mode - iterative-retrieval
packages/squad-cli/templates/skills/iterative-retrieval - model-selection
packages/squad-cli/templates/skills/model-selection - nap
packages/squad-cli/templates/skills/nap - notification-routing
packages/squad-cli/templates/skills/notification-routing - pr-review-response
packages/squad-cli/templates/skills/pr-review-response - pr-screenshots
packages/squad-cli/templates/skills/pr-screenshots - project-conventions
packages/squad-cli/templates/skills/project-conventions - ralph-two-pass-scan
packages/squad-cli/templates/skills/ralph-two-pass-scan - reflect
packages/squad-cli/templates/skills/reflect - release-process
packages/squad-cli/templates/skills/release-process - reskill
packages/squad-cli/templates/skills/reskill - retro-enforcement
packages/squad-cli/templates/skills/retro-enforcement - reviewer-protocol
packages/squad-cli/templates/skills/reviewer-protocol - secret-handling
packages/squad-cli/templates/skills/secret-handling - session-recovery
packages/squad-cli/templates/skills/session-recovery - squad
packages/squad-cli/templates/skills/squad - squad-commands
packages/squad-cli/templates/skills/squad-commands - squad-conventions
packages/squad-cli/templates/skills/squad-conventions - squad-help
packages/squad-cli/templates/skills/squad-help - squad-version-check
packages/squad-cli/templates/skills/squad-version-check - test-discipline
packages/squad-cli/templates/skills/test-discipline - tiered-memory
packages/squad-cli/templates/skills/tiered-memory - versioning-policy
packages/squad-cli/templates/skills/versioning-policy - windows-compatibility
packages/squad-cli/templates/skills/windows-compatibility - {skill-name}
packages/squad-sdk/templates - agent-collaboration
packages/squad-sdk/templates/skills/agent-collaboration - agent-conduct
packages/squad-sdk/templates/skills/agent-conduct - architectural-proposals
packages/squad-sdk/templates/skills/architectural-proposals - ci-validation-gates
packages/squad-sdk/templates/skills/ci-validation-gates - cli-wiring
packages/squad-sdk/templates/skills/cli-wiring - client-compatibility
packages/squad-sdk/templates/skills/client-compatibility - coordinator-init-mode
packages/squad-sdk/templates/skills/coordinator-init-mode - coordinator-response-mode
packages/squad-sdk/templates/skills/coordinator-response-mode - coordinator-source-of-truth
packages/squad-sdk/templates/skills/coordinator-source-of-truth - cross-machine-coordination
packages/squad-sdk/templates/skills/cross-machine-coordination - cross-squad
packages/squad-sdk/templates/skills/cross-squad - cross-squad-communication
packages/squad-sdk/templates/skills/cross-squad-communication - distributed-mesh
packages/squad-sdk/templates/skills/distributed-mesh - docs-standards
packages/squad-sdk/templates/skills/docs-standards - e2e-template-testing
packages/squad-sdk/templates/skills/e2e-template-testing - economy-mode
packages/squad-sdk/templates/skills/economy-mode - error-recovery
packages/squad-sdk/templates/skills/error-recovery - external-comms
packages/squad-sdk/templates/skills/external-comms - fact-checking
packages/squad-sdk/templates/skills/fact-checking - gh-auth-isolation
packages/squad-sdk/templates/skills/gh-auth-isolation - gh-aw-enlistment
packages/squad-sdk/templates/skills/gh-aw-enlistment - git-workflow
packages/squad-sdk/templates/skills/git-workflow - github-multi-account
packages/squad-sdk/templates/skills/github-multi-account - history-hygiene
packages/squad-sdk/templates/skills/history-hygiene - humanizer
packages/squad-sdk/templates/skills/humanizer - init-mode
packages/squad-sdk/templates/skills/init-mode - iterative-retrieval
packages/squad-sdk/templates/skills/iterative-retrieval - model-selection
packages/squad-sdk/templates/skills/model-selection - nap
packages/squad-sdk/templates/skills/nap - notification-routing
packages/squad-sdk/templates/skills/notification-routing - pr-review-response
packages/squad-sdk/templates/skills/pr-review-response - pr-screenshots
packages/squad-sdk/templates/skills/pr-screenshots - project-conventions
packages/squad-sdk/templates/skills/project-conventions - ralph-two-pass-scan
packages/squad-sdk/templates/skills/ralph-two-pass-scan - reflect
packages/squad-sdk/templates/skills/reflect - release-process
packages/squad-sdk/templates/skills/release-process - reskill
packages/squad-sdk/templates/skills/reskill - retro-enforcement
packages/squad-sdk/templates/skills/retro-enforcement - reviewer-protocol
packages/squad-sdk/templates/skills/reviewer-protocol - secret-handling
packages/squad-sdk/templates/skills/secret-handling - session-recovery
packages/squad-sdk/templates/skills/session-recovery - squad
packages/squad-sdk/templates/skills/squad - squad-commands
packages/squad-sdk/templates/skills/squad-commands - squad-conventions
packages/squad-sdk/templates/skills/squad-conventions - squad-help
packages/squad-sdk/templates/skills/squad-help - squad-version-check
packages/squad-sdk/templates/skills/squad-version-check - test-discipline
packages/squad-sdk/templates/skills/test-discipline - tiered-memory
packages/squad-sdk/templates/skills/tiered-memory - versioning-policy
packages/squad-sdk/templates/skills/versioning-policy - windows-compatibility
packages/squad-sdk/templates/skills/windows-compatibility - {skill-name}
templates - agent-collaboration
templates/skills/agent-collaboration - agent-conduct
templates/skills/agent-conduct - architectural-proposals
templates/skills/architectural-proposals - ci-validation-gates
templates/skills/ci-validation-gates - cli-wiring
templates/skills/cli-wiring - client-compatibility
templates/skills/client-compatibility - coordinator-init-mode
templates/skills/coordinator-init-mode - coordinator-response-mode
templates/skills/coordinator-response-mode - coordinator-source-of-truth
templates/skills/coordinator-source-of-truth - cross-machine-coordination
templates/skills/cross-machine-coordination - cross-squad
templates/skills/cross-squad - cross-squad-communication
templates/skills/cross-squad-communication - distributed-mesh
templates/skills/distributed-mesh - docs-standards
templates/skills/docs-standards - e2e-template-testing
templates/skills/e2e-template-testing - economy-mode
templates/skills/economy-mode - error-recovery
templates/skills/error-recovery - external-comms
templates/skills/external-comms - fact-checking
templates/skills/fact-checking - gh-auth-isolation
templates/skills/gh-auth-isolation - gh-aw-enlistment
templates/skills/gh-aw-enlistment - git-workflow
templates/skills/git-workflow - github-multi-account
templates/skills/github-multi-account - history-hygiene
templates/skills/history-hygiene - humanizer
templates/skills/humanizer - init-mode
templates/skills/init-mode - iterative-retrieval
templates/skills/iterative-retrieval - model-selection
templates/skills/model-selection - nap
templates/skills/nap - notification-routing
templates/skills/notification-routing - pr-review-response
templates/skills/pr-review-response - pr-screenshots
templates/skills/pr-screenshots - project-conventions
templates/skills/project-conventions - ralph-two-pass-scan
templates/skills/ralph-two-pass-scan - reflect
templates/skills/reflect - release-process
templates/skills/release-process - reskill
templates/skills/reskill - retro-enforcement
templates/skills/retro-enforcement - reviewer-protocol
templates/skills/reviewer-protocol - secret-handling
templates/skills/secret-handling - session-recovery
templates/skills/session-recovery - squad
templates/skills/squad - squad-commands
templates/skills/squad-commands - squad-conventions
templates/skills/squad-conventions - squad-help
templates/skills/squad-help - squad-version-check
templates/skills/squad-version-check - test-discipline
templates/skills/test-discipline - tiered-memory
templates/skills/tiered-memory - versioning-policy
templates/skills/versioning-policy - windows-compatibility
templates/skills/windows-compatibility - bad-skill
test-fixtures/skills-test-abb30454/bad-skill - My Skill
test-fixtures/skills-test-abb30454/my-skill - gh-aw-enlistment
workflows/skills/gh-aw-enlistment
How it works
action-skill-contracts
When your agent uses it
Author deterministic skills for operations that change state or invoke external tools
What it loads into context
Frontmatter
48 tokensAlways available to the agent
SKILL.md
328 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,725 bytes
Read SKILL.md
## Context
Use this skill when authoring or revising a skill that can mutate files, state, repositories, or
external systems. Informational reference skills need not carry operational ceremony.
## Patterns
- Start with an explicit `## SCOPE`: activation conditions, allowed targets, and exclusions.
- Provide an ordered, deterministic workflow. State required inputs, the intended side effect, and
the exact validation after each consequential operation.
- Define STOP conditions for missing prerequisites, ambiguous scope, failed verification, and
unavailable tools. A failed precondition must not look like success.
- Point to a current canonical implementation or workflow when a procedure depends on repository
behavior; do not copy volatile values into prose.agent-collaboration
When your agent uses it
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
440 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore2,010 bytes
Read SKILL.md
## Context
Every agent on the team follows identical collaboration patterns for worktree awareness, decision recording, and cross-agent communication. These were previously duplicated in every charter's Collaboration section (~300 bytes × 18 agents = ~5.4KB of redundant context). Now centralized here.
The coordinator's spawn prompt already instructs agents to read decisions.md and their history.md. This skill adds the patterns for WRITING decisions and requesting help.
## Patterns
### Worktree Awareness
Use the `TEAM ROOT` path provided in your spawn prompt. All `.squad/` paths are relative to this root. If TEAM ROOT is not provided (rare), run `git rev-parse --show-toplevel` as fallback. Never assume CWD is the repo root.
### Decision Recording
After making a decision that affects other team members, write it to:
`.squad/decisions/inbox/{your-name}-{brief-slug}.md`agent-conduct
When your agent uses it
Shared hard rules enforced across all squad agents
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
293 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore1,359 bytes
Read SKILL.md
## Context
Every squad agent must follow these two hard rules. They were previously duplicated in every charter. Now they live here as a shared skill, loaded once.
## Patterns
### Product Isolation Rule (hard rule)
Tests, CI workflows, and product code must NEVER depend on specific agent names from any particular squad. "Our squad" must not impact "the squad." No hardcoded references to agent names (Flight, EECOM, FIDO, etc.) in test assertions, CI configs, or product logic. Use generic/parameterized values. If a test needs agent names, use obviously-fake test fixtures (e.g., "test-agent-1", "TestBot").
### Peer Quality Check (hard rule)
Before finishing work, verify your changes don't break existing tests. Run the test suite for files you touched. If CI has been failing, check your changes aren't contributing to the problem. When you learn from mistakes, update your history.md.
## Anti-Patterns
- Don't hardcode dev team agent names in product code or testsarchitectural-proposals
When your agent uses it
How to write comprehensive architectural proposals that drive alignment before code is written
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
1,430 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,343 bytes
Read SKILL.md
## Context
Proposals create alignment before code is written. Cheaper to change a doc than refactor code. Use this pattern when:
- Architecture shifts invalidate existing assumptions
- Product direction changes require new foundation
- Multiple waves/milestones will be affected by a decision
- External dependencies (Copilot CLI, SDK APIs) change
## Patterns
### Proposal Structure (docs/proposals/)
**Required sections:**
1. **Problem Statement** — Why current state is broken (specific, measurable evidence)architectural-review
When your agent uses it
How to review PRs for architectural quality — module boundaries, dependency direction, export surface, pattern consistency
What it loads into context
Frontmatter
58 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Review Output Format
- Anti-Patterns
Files
SKILL.mdcore9,036 bytes
Read SKILL.md
## Context
When reviewing a PR that touches core architecture, the reviewer must verify that structural invariants are preserved. This skill covers **reviewing existing PRs** for architectural quality — not writing proposals (see `architectural-proposals/SKILL.md` for that).
Use this skill when a PR includes any of:
- New modules, packages, or directories
- Dependency changes (`package.json`, `import` additions across package boundaries)
- API surface changes (new or modified exports from `index.ts` barrel files)
- Cross-package imports (anything crossing the SDK/CLI boundary)
- New abstractions (interfaces, base classes, registries, factories)
- Sweeping refactors (file renames, pattern migrations, bulk conversions)
## Patterns
archival-integrity
When your agent uses it
Preserve content when moving entries between tracked Squad state files
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
591 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Patterns
- Stop conditions
- Examples
- Anti-Patterns
Files
SKILL.mdcore2,944 bytesSKILL.mdcore2,944 bytesSKILL.mdcore2,944 bytesSKILL.mdcore2,944 bytes
Read SKILL.md
## Context
Use this skill for decision archival, history summarization, or any operation that moves entries
between repository state files. A move has two halves: append to the destination, then trim the
source. Treat it as data loss prevention, not text cleanup.
## Scope
- Activate when moving entries between tracked Squad state files such as decision logs, histories, or inbox merges.
- Allowed targets: tracked repository state files and their designated tracked archives.
- Exclusions: untracked destinations, byte-delta-only validation, and trimming source content before the destination is verified.
- This procedure is scoped to the local state backend unless the active backend exposes governed
append, read-back, and trim operations. Runtime-owned orphan or two-layer state is not a valid
archival destination; use the backend's state operation instead of editing a checkout copy.ci-validation-gates
When your agent uses it
Defensive CI/CD patterns: semver validation, token checks, retry logic, and draft detection
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,056 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Known Failure Modes
- Anti-Patterns
Scripts executed separately
../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,478 bytes../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
CI workflows must be defensive. These patterns capture lessons from prior release incidents;
they are maintained as version-agnostic gates rather than claims about a particular repository
release.
## Patterns
### Semver Validation Gate
Every publish workflow MUST validate version format before `npm publish`. 4-part versions (e.g., 0.8.21.4) are NOT valid semver — npm mangles them.
```yaml
- name: Validate semver
run: |cli-wiring
When your agent uses it
Checklist and patterns for wiring new CLI commands into cli-entry.ts
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
532 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Checklist — Adding a New CLI Command
- Wiring Patterns by Command Type
- Common Import Pattern
- History
Files
SKILL.mdcore2,009 bytes
Read SKILL.md
# Skill: CLI Command Wiring
**Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.
## Checklist — Adding a New CLI Command
1. **Create command file** in `packages/squad-cli/src/cli/commands/<name>.ts`
- Export a `run<Name>(cwd, options)` async function (or class with static methods for utility modules)
2. **Add routing block** in `packages/squad-cli/src/cli-entry.ts` inside `main()`:
```ts
if (cmd === '<name>') {
const { run<Name> } = await import('./cli/commands/<name>.js');
// parse args, call functionclient-compatibility
When your agent uses it
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,289 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,719 bytes
Read SKILL.md
## Context
Squad runs on multiple Copilot surfaces (CLI, VS Code, JetBrains, GitHub.com). The coordinator must detect its platform and adapt spawning behavior accordingly. Different tools are available on different platforms, requiring conditional logic for agent spawning, SQL usage, and response timing.
## Patterns
### Platform Detection
Before spawning agents, determine the platform by checking available tools:
1. **CLI mode** — `task` tool is available → full spawning control. Use `task` with `agent_type`, `mode`, `model`, `description`, `prompt` parameters. Collect results via `read_agent`.
2. **VS Code mode** — `runSubagent` or `agent` tool is available → conditional behavior. Use `runSubagent` with the task prompt. Drop `agent_type`, `mode`, and `model` parameters. Multiple subagents in one turn run concurrently (equivalent to background mode). Results return automatically — no `read_agent` needed.
content-scope
When your agent uses it
Keep Squad documentation focused on shipped behavior and clearly bounded integrations
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
249 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,420 bytes
Read SKILL.md
## Context
Use this skill when deciding whether proposed documentation, examples, or external material
belongs in Squad documentation.
## Patterns
- Apply the “Squad ships it” test: keep behavior, configuration, and integration contracts that
Squad ships; exclude external infrastructure that users build around Squad.
- When documenting a platform feature used with Squad, identify it as a platform integration
rather than implying Squad implements the platform feature.
- Preserve reusable product guidance while linking to external examples, articles, or deployment
material instead of importing their operational specifics into product documentation.
distributed-mesh
When your agent uses it
How to coordinate with squads on different machines using git as transport
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
3,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Patterns
- Examples
- AGENT WORKFLOW (Deterministic Setup)
- Anti-Patterns
Scripts executed separately
sync-mesh.ps1sync-mesh.sh
Files
SKILL.mdcore12,433 bytesSKILL.mdcore12,433 bytesSKILL.mdcore12,433 bytesSKILL.mdcore12,433 bytesSKILL.mdcore12,433 bytesmesh.json.exampleasset762 bytessync-mesh.ps1runs3,885 bytessync-mesh.shruns3,409 bytes
Read SKILL.md
## SCOPE
**✅ THIS SKILL PRODUCES (exactly these, nothing more):**
1. **`mesh.json`** — Generated from user answers about zones and squads (which squads participate, what zone each is in, paths/URLs for each), using `mesh.json.example` in this skill's directory as the schema template
2. **`sync-mesh.sh` and `sync-mesh.ps1`** — Copied from this skill's directory into the project root (these are bundled resources, NOT generated code)
3. **Zone 2 state repo initialization** (if applicable) — If the user specified a Zone 2 shared state repo, run `sync-mesh.sh --init` to scaffold the state repo structure
4. **A decision entry** in `.squad/decisions/inbox/` documenting the mesh configuration for team awareness
**❌ THIS SKILL DOES NOT PRODUCE:**
- **No application code** — No validators, libraries, or modules of any kind
- **No test files** — No test suites, test cases, or test scaffolding
- **No GENERATING sync scripts** — They are bundled with this skill as pre-built resources. COPY them, don't generate them.documentation-maintenance
When your agent uses it
Maintain accurate, scannable Squad documentation with validated inventory and links
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
319 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,698 bytes
Read SKILL.md
## Context
Use this skill when adding, removing, or materially changing user-facing documentation. Squad
documentation should describe the product and its supported configuration, not surrounding
infrastructure presented as product behavior.
## Patterns
- Follow the Microsoft Style Guide: sentence-case headings, active voice, second person, and
present tense unless established project voice requires an exception.
- Keep one canonical page per concept. Link to it rather than duplicating content, using the most
specific stable anchor available.
- Choose paragraphs for explanation, parallel lists for scannable choices or steps, and tables for
comparisons or reference data.git-workflow
When your agent uses it
Squad branching model: dev-first workflow with insiders preview channel
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,585 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Branch Naming Convention
- Workflow for Issue Work
- Parallel Multi-Issue Work (Worktrees)
- Multi-Repo Downstream Scenarios
- Anti-Patterns
- Promotion Pipeline
Files
SKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytesSKILL.mdcore6,255 bytes
Read SKILL.md
## Context
Squad uses a three-branch model. **All feature work starts from `dev`, not `main`.**
| Branch | Purpose | Publishes |
|--------|---------|-----------|
| `main` | Released, tagged, in-npm code only | `npm publish` on tag |
| `dev` | Integration branch — all feature work lands here | `npm publish --tag preview` on merge |
| `insiders` | Early-access channel — synced from dev | `npm publish --tag insiders` on sync |
## Branch Naming Convention
Issue branches MUST use: `squad/{issue-number}-{kebab-case-slug}`
github-multi-account
When your agent uses it
Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
What it loads into context
Frontmatter
80 tokensAlways available to the agent
SKILL.md
1,038 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Activate
- What to Do (as the AI agent)
- After Setup — Usage Rules
- Repo-Specific Account Binding
- For Squad Agents
Files
SKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytes
Read SKILL.md
# GitHub Multi-Account — AI-Driven Setup
## When to Activate
When the user has multiple GitHub accounts (check with `gh auth status`). If you see 2+ accounts listed, this skill applies.
## What to Do (as the AI agent)
### Step 1: Detect accounts
Run: `gh auth status`
Look for multiple accounts. Note which usernames are listed.
### Step 2: Ask the user
Ask: "I see you have multiple GitHub accounts: {list them}. Which one is your personal account and which is your work/EMU account?"
history-hygiene
When your agent uses it
Record reconciled final outcomes to history.md, not intermediate requests or obsolete claims
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
391 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,905 bytesSKILL.mdcore1,905 bytesSKILL.mdcore1,905 bytesSKILL.mdcore1,905 bytes
Read SKILL.md
## Context
History files are cold-start context, not the source of truth for live behavior. Stale or incorrect
entries poison decision-making downstream. A stale-history incident proved that an old history claim
can cause future spawns to repeat a reversed decision. Reconcile an operational claim with current
implementation, workflows, tests, documentation, skills, and applicable decisions before acting on it.
## Patterns
- **Record the final outcome**, not the initial request.
- **Wait for confirmation** before writing to history — don't log intermediate states.
- **If a decision reverses**, update the entry immediately — don't leave stale data.
- **Link authority for operational facts.** Name the current source or workflow when a lesson
depends on live behavior.init-mode
When your agent uses it
Team initialization flow (Phase 1 proposal + Phase 2 creation)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
1,686 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,768 bytes
Read SKILL.md
## Context
Init Mode activates when `.squad/team.md` does not exist, or exists but has zero roster entries under `## Members`. The coordinator proposes a team (Phase 1), waits for user confirmation, then creates the team structure (Phase 2).
## Patterns
### Phase 1: Propose the Team
No team exists yet. Propose one — but **DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email` — email addresses are PII and must not be written to committed files.**
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see that section):
- Determine team size (typically 4–5 + Scribe).model-selection
When your agent uses it
Per-agent model selection with 4-layer hierarchy and fallback chains
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
2,188 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore8,047 bytes
Read SKILL.md
## Context
Before spawning an agent, the coordinator determines which model to use. This skill codifies the 4-layer hierarchy, role-to-model mappings, task complexity adjustments, and fallback chains. Applies to all agent spawns in Team Mode.
## Patterns
### 4-Layer Hierarchy
Check these layers in order — first match wins:
**Layer 1 — User Override:** Did the user specify a model? ("use opus", "save costs", "use gpt-5.3-codex for this"). If yes, use that model. Session-wide directives ("always use haiku") persist until contradicted.
**Layer 2 — Charter Preference:** Does the agent's charter have a `## Model` section with `Preferred` set to a specific model (not `auto`)? If yes, use that model.
pr-lifecycle
When your agent uses it
Complete issue → PR → merge lifecycle with readiness checks
What it loads into context
Frontmatter
50 tokensAlways available to the agent
SKILL.md
5,030 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Lifecycle Procedure
- Examples
- Anti-Patterns
- Readiness Check Gaps & Recommendations
Files
SKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytesSKILL.mdcore19,249 bytes
Read SKILL.md
## Context
This skill is the **canonical Copilot-agent lifecycle for the Squad repository**. It covers the full path from picking up a GitHub issue to merging a PR. Where older docs (templates, copilot-instructions, CONTRIBUTING.md) conflict with this skill, **this skill takes precedence** for Copilot agents.
For advanced scenarios (worktrees, multi-repo coordination), see `.copilot/skills/git-workflow/SKILL.md`.
## Scope
✅ **THIS SKILL COVERS:**
- Issue pickup and branch creation
- Implementation, build, test, lint workflow
- Pre-push safety verification
- PR creation with correct target, title, body, and labels
- All 11 PR readiness checks (what they check, how to pass, how to fix)protected-files
When your agent uses it
Zero-dependency bootstrap files that must never import npm packages or SDK code
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
695 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Protected File List
- Rules
- SDK/CLI Package Boundary
- Anti-Patterns
- Adding New Bootstrap Utilities
Files
SKILL.mdcore2,873 bytes
Read SKILL.md
## Context
The CLI (`squad-cli`) has bootstrap utilities that run **before** the Squad SDK is loaded. If these files import SDK code (e.g., `FSStorageProvider`, anything from `squad-sdk`), the CLI breaks at startup — no helpful error, just a crash.
This skill applies when:
- Touching any file in `packages/squad-cli/src/cli/core/`
- Running sweeping refactors (e.g., "convert all `fs` calls to `StorageProvider`")
- Adding new bootstrap utilities
## Protected File List
| File | Purpose |
|------|---------|
| `packages/squad-cli/src/cli/core/detect-squad-dir.ts` | Finds `.squad/` directory at startup — runs before SDK init |release-process
When your agent uses it
Operate Squad's automated insider, preview, and stable release channels safely
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,618 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Preconditions
- Preview
- Insider
- Stable
- Verify
- Recovery
Files
SKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytesSKILL.mdcore6,725 bytes
Read SKILL.md
# Squad Release Process
Read `.squad/skills/release-process/SKILL.md` for the canonical runbook and
recovery commands. The operational model is:
| Release | Trigger | Required version | Result |
|---------|---------|------------------|--------|
| Insider | Manual `squad-insider-publish.yml` dispatch from `dev` | Generated `X.Y.Z-insider.N` | GitHub prerelease, npm `insider`, standalone archives, Homebrew, WinGet |
| Preview | Manual `squad-release.yml` dispatch from `dev` | `X.Y.Z-preview.N` | GitHub prerelease, npm `preview`, standalone archives, Homebrew, WinGet |
| Stable | Manual `squad-promote.yml` dispatch from `dev` | `X.Y.Z` | Sanitized promotion PR or legacy `main` push, then GitHub stable release, npm `latest`, standalone archives, Homebrew, WinGet |
There is no staging `preview` branch. Do not create tags or GitHub Releases
manually.
reskill
When your agent uses it
Extract durable operating knowledge before slimming or replacing team charters
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,223 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Process
- Patterns
- Anti-Patterns
Files
SKILL.mdcore5,760 bytesSKILL.mdcore5,760 bytes
Read SKILL.md
## Context
When the coordinator hears "team, reskill" (or similar: "optimize context", "slim down charters"), trigger a team-wide optimization pass. The goal: reduce per-agent context consumption by extracting shared patterns from charters and histories into reusable skills.
This is a periodic maintenance activity. Run whenever charter/history bloat is suspected or a
team will be re-cast. Charters and histories are evidence, not the authority for current product
behavior: reconcile candidates against current source, workflows, tests, documentation, skills,
and decisions before preserving them.
## Process
### Step 1: Audit
Read all active agent charters and histories, then inventory existing skills. Measure byte sizes.
Identify:reviewer-protocol
When your agent uses it
Reviewer rejection workflow and strict lockout semantics
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
1,009 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,658 bytes
Read SKILL.md
## Context
When a team member has a **Reviewer** role (e.g., Tester, Code Reviewer, Lead), they may approve or reject work from other agents. On rejection, the coordinator enforces strict lockout rules to ensure the original author does NOT self-revise. This prevents defensive feedback loops and ensures independent review.
## Patterns
### Reviewer Rejection Protocol
Quality and test reviewers may block a PR or merge on quality, coverage, or regression grounds. Treat that block as a rejection that enters the protocol below.
When a team member has a **Reviewer** role:
- Reviewers may **approve** or **reject** work from other agents.
- On **rejection**, the Reviewer may choose ONE of:secret-handling
When your agent uses it
Never read .env files or write secrets to .squad/ committed files
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
1,998 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,547 bytesSKILL.mdcore7,547 bytes
Read SKILL.md
## Context
Spawned agents have read access to the entire repository, including `.env` files containing live credentials. If an agent reads secrets and writes them to `.squad/` files (decisions, logs, history), Scribe auto-commits them to git, exposing them in remote history. This skill codifies absolute prohibitions and safe alternatives.
## Patterns
### Prohibited File Reads
**NEVER read these files:**
- `.env` (production secrets)
- `.env.local` (local dev secrets)
- `.env.production` (production environment)
- `.env.development` (development environment)
- `.env.staging` (staging environment)security-review
When your agent uses it
How to review PRs for security — credentials, injection, workflow permissions, supply chain, git operation safety
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
2,669 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Review Output Format
- Anti-Patterns
Files
SKILL.mdcore11,732 bytesSKILL.mdcore11,732 bytesSKILL.mdcore11,732 bytes
Read SKILL.md
## Context
Every PR that touches authentication, credentials, environment variables, file system operations, child process execution, user input handling, GitHub API calls, or workflow files must be reviewed for security. This skill provides a systematic checklist for catching security issues before they reach production.
Use this skill when a PR includes any of:
- Changes to `.github/workflows/` files
- New or modified environment variable usage
- File system read/write operations driven by user input
- Child process execution (`child_process.exec`, `child_process.spawn`)
- GitHub API calls (Octokit, `gh` CLI wrappers)
- New npm dependencies
- Credential or token handling
- Template rendering with dynamic values
squad-conventions
When your agent uses it
Core conventions and patterns used in the Squad codebase
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
824 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore3,442 bytesSKILL.mdcore3,442 bytes
Read SKILL.md
## Context
These conventions apply to all work on the Squad CLI tool (`create-squad`). Squad is a zero-dependency Node.js package that adds AI agent teams to any project. Understanding these patterns is essential before modifying any Squad source code.
## Patterns
### Zero Dependencies
Squad has zero runtime dependencies. Everything uses Node.js built-ins (`fs`, `path`, `os`, `child_process`). Do not add packages to `dependencies` in `package.json`. This is a hard constraint, not a preference.
### Node.js Built-in Test Runner
Tests use `node:test` and `node:assert/strict` — no test frameworks. Run with `npm test`. Test files live in `test/`. The test command is `node --test test/`.
### Error Handling — `fatal()` Pattern
All user-facing errors use the `fatal(msg)` function which prints a red `✗` prefix and exits with code 1. Never throw unhandled exceptions or print raw stack traces. The global `uncaughtException` handler calls `fatal()` as a safety net.
test-discipline
When your agent uses it
Design assertions that prove the actual contract and fail on realistic regressions
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytes
Read SKILL.md
## Context
Use this skill whenever a change affects an API, public interface, workflow contract, generated
artifact, or user-observable behavior. A passing status-only test is not evidence that the
intended artifact or contract is correct.
## Patterns
- Update tests in the same change as an API, signature, export, or public-interface change.
- Make assertions name and inspect the offending artifact, output, or configuration field—not just
an exit code or broad success status.
- Mutation-test critical gates by changing the real source or emitted artifact and proving the
assertion turns red.
- Exercise every caller path for a changed parser, normalizer, or helper; isolated unit coveragewindows-compatibility
When your agent uses it
Cross-platform path handling and command patterns
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
1,572 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../scripts/fix-crlf-worktree.mjs
Files
SKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytes../../../scripts/fix-crlf-worktree.mjsruns5,444 bytes
Read SKILL.md
## Context
Squad runs on Windows, macOS, and Linux. Several bugs have been traced to platform-specific assumptions: ISO timestamps with colons (illegal on Windows), `git -C` with Windows paths (unreliable), forward-slash paths in Node.js on Windows.
## Patterns
### Filenames & Timestamps
- **Never use colons in filenames:** ISO 8601 format `2026-03-15T05:30:00Z` is illegal on Windows
- **Use `safeTimestamp()` utility:** Replaces colons with hyphens → `2026-03-15T05-30-00Z`
- **Centralize formatting:** Don't inline `.toISOString().replace(/:/g, '-')` — use the utility
### Git Commands
- **Never use `git -C {path}`:** Unreliable with Windows paths (backslashes, spaces, drive letters)
- **Always `cd` first:** Change directory, then run git commands{skill-name}
When your agent uses it
{what this skill teaches agents}
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
145 tokensWhen the skill is activated
Reference documents
66,826 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
package.jsonralph-triage.jsscripts/notes/fetch.ps1scripts/notes/write-note.ps1
Files
README.mddoc7,063 bytesafter-agent-reference.mddoc5,750 bytescasting-history.jsonasset70 bytescasting-policy.jsonasset823 bytescasting-reference.mddoc7,679 bytescasting-registry.jsonasset146 bytescasting/Futurama.jsonasset108 bytesceremonies.mddoc1,882 bytesceremony-reference.mddoc3,162 bytescharter.mddoc2,053 bytesclient-compatibility-reference.mddoc3,758 bytesconstraint-tracking.mddoc1,192 bytescooperative-rate-limiting.mddoc7,222 bytescopilot-agent.mddoc3,777 bytescopilot-instructions.mddoc3,667 bytesfact-checker-charter.mddoc2,621 bytesfact-checker-policy.mddoc6,104 byteshistory.mddoc255 bytesidentity/now.mddoc209 bytesidentity/wisdom.mddoc384 bytesissue-lifecycle.mddoc11,524 byteskeda-scaler.mddoc5,732 bytesmachine-capabilities.mddoc2,415 bytesmcp-config.mddoc3,129 bytesmodel-selection-reference.mddoc16,277 bytesmulti-agent-format.mddoc887 bytesnotes-protocol.mddoc5,513 bytesorchestration-log.mddoc1,104 bytespackage.jsonruns28 bytesplugin-marketplace.mddoc2,437 bytesprd-intake.mddoc3,311 bytesrai-charter.mddoc2,941 bytesrai-policy.mddoc4,057 bytesralph-charter.mddoc2,236 bytesralph-circuit-breaker.mddoc10,439 bytesralph-instructions.mddoc3,445 bytesralph-reference.mddoc8,828 bytesralph-triage.jsruns16,280 bytesraw-agent-output.mddoc1,336 bytesroster.mddoc2,389 bytesrouting.mddoc1,789 bytesrun-output.mddoc1,500 bytesscribe-charter.mddoc3,001 bytesscripts/notes/fetch.ps1runs3,977 bytesscripts/notes/write-note.ps1runs5,449 bytessession-init-reference.mddoc7,194 bytesskill.mdcore587 bytesspawn-reference.mddoc9,351 bytessquad.agent.mddoc80,210 bytesworkflow-wiring-appendix-a-code-reviewer.mddoc6,323 bytesworkflow-wiring-appendix-b-documenter.mddoc6,349 bytesworkflow-wiring-guide.mddoc14,509 bytesworkflows/squad-ci.ymlasset506 bytesworkflows/squad-docs.ymlasset1,264 bytesworkflows/squad-heartbeat.ymlasset4,857 bytesworkflows/squad-insider-release.ymlasset2,307 bytesworkflows/squad-issue-assign.ymlasset9,838 bytesworkflows/squad-label-enforce.ymlasset6,935 bytesworkflows/squad-preview.ymlasset1,842 bytesworkflows/squad-promote.ymlasset4,363 bytesworkflows/squad-release.ymlasset2,722 bytesworkflows/squad-triage.ymlasset14,752 bytesworkflows/sync-squad-labels.ymlasset9,408 bytesworktree-reference.mddoc7,553 bytes
Read SKILL.md
## Context
{When and why this skill applies}
## Patterns
{Specific patterns, conventions, or approaches}
## Examples
{Code examples or references}
## Anti-Patterns
{What to avoid}agent-collaboration
When your agent uses it
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
440 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore2,010 bytes
Read SKILL.md
## Context
Every agent on the team follows identical collaboration patterns for worktree awareness, decision recording, and cross-agent communication. These were previously duplicated in every charter's Collaboration section (~300 bytes × 18 agents = ~5.4KB of redundant context). Now centralized here.
The coordinator's spawn prompt already instructs agents to read decisions.md and their history.md. This skill adds the patterns for WRITING decisions and requesting help.
## Patterns
### Worktree Awareness
Use the `TEAM ROOT` path provided in your spawn prompt. All `.squad/` paths are relative to this root. If TEAM ROOT is not provided (rare), run `git rev-parse --show-toplevel` as fallback. Never assume CWD is the repo root.
### Decision Recording
After making a decision that affects other team members, write it to:
`.squad/decisions/inbox/{your-name}-{brief-slug}.md`agent-conduct
When your agent uses it
Shared hard rules enforced across all squad agents
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
283 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore1,345 bytes
Read SKILL.md
## Context
Every squad agent must follow these two hard rules. They were previously duplicated in every charter. Now they live here as a shared skill, loaded once.
## Patterns
### Product Isolation Rule (hard rule)
Tests, CI workflows, and product code must NEVER depend on specific agent names from any particular squad. "Our squad" must not impact "the squad." No hardcoded references to cast-specific agent names in test assertions, CI configs, or product logic. Use generic/parameterized values. If a test needs agent names, use obviously-fake test fixtures (e.g., "test-agent-1", "TestBot").
### Peer Quality Check (hard rule)
Before finishing work, verify your changes don't break existing tests. Run the test suite for files you touched. If CI has been failing, check your changes aren't contributing to the problem. When you learn from mistakes, update your history.md.
## Anti-Patterns
- Don't hardcode dev team agent names in product code or testsarchitectural-proposals
When your agent uses it
How to write comprehensive architectural proposals that drive alignment before code is written
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
1,427 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,335 bytes
Read SKILL.md
## Context
Proposals create alignment before code is written. Cheaper to change a doc than refactor code. Use this pattern when:
- Architecture shifts invalidate existing assumptions
- Product direction changes require new foundation
- Multiple waves/milestones will be affected by a decision
- External dependencies (Copilot CLI, SDK APIs) change
## Patterns
### Proposal Structure (docs/proposals/)
**Required sections:**
1. **Problem Statement** — Why current state is broken (specific, measurable evidence)ci-validation-gates
When your agent uses it
Defensive CI/CD patterns: semver validation, token checks, retry logic, and draft detection
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,072 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Known Failure Modes
- Anti-Patterns
Scripts executed separately
../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,562 bytes../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
CI workflows must be defensive. These patterns capture lessons from prior release incidents;
they are maintained as version-agnostic gates rather than claims about a particular repository
release.
## Patterns
### Semver Validation Gate
Every publish workflow MUST validate version format before `npm publish`. 4-part versions (e.g., 0.8.21.4) are NOT valid semver — npm mangles them.
```yaml
- name: Validate semver
run: |When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
491 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Checklist — Adding a New CLI Command
- Wiring Patterns by Command Type
- Common Import Pattern
- History
Files
SKILL.mdcore1,844 bytes
Read SKILL.md
# Skill: CLI Command Wiring
**Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.
## Checklist — Adding a New CLI Command
1. **Create command file** in `packages/squad-cli/src/cli/commands/<name>.ts`
- Export a `run<Name>(cwd, options)` async function (or class with static methods for utility modules)
2. **Add routing block** in `packages/squad-cli/src/cli-entry.ts` inside `main()`:
```ts
if (cmd === '<name>') {
const { run<Name> } = await import('./cli/commands/<name>.js');
// parse args, call function
await run<Name>(process.cwd(), options);client-compatibility
When your agent uses it
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,289 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,719 bytes
Read SKILL.md
## Context
Squad runs on multiple Copilot surfaces (CLI, VS Code, JetBrains, GitHub.com). The coordinator must detect its platform and adapt spawning behavior accordingly. Different tools are available on different platforms, requiring conditional logic for agent spawning, SQL usage, and response timing.
## Patterns
### Platform Detection
Before spawning agents, determine the platform by checking available tools:
1. **CLI mode** — `task` tool is available → full spawning control. Use `task` with `agent_type`, `mode`, `model`, `description`, `prompt` parameters. Collect results via `read_agent`.
2. **VS Code mode** — `runSubagent` or `agent` tool is available → conditional behavior. Use `runSubagent` with the task prompt. Drop `agent_type`, `mode`, and `model` parameters. Multiple subagents in one turn run concurrently (equivalent to background mode). Results return automatically — no `read_agent` needed.
coordinator-init-mode
When your agent uses it
The complete two-phase Init Mode protocol the Squad coordinator runs when no team exists yet in the current repo. Phase 1 = propose the team (no files created, wait for user confirm). Phase 2 = create .squad/ scaffolding, casting state, .gitattributes for merge drivers, and the always-on built-ins (Scribe, Ralph, Rai, Fact Checker). Loaded on demand when the coordinator detects no .squad/team.md exists.
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
1,937 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Phase 1: Propose the Team
- Phase 2: Create the Team
Files
SKILL.mdcore7,868 bytes
Read SKILL.md
> **Load this skill when:** you detect that no `.squad/team.md` exists in the resolved team root — that means this is a fresh repo or a repo that has never been squadified, and Init Mode is the right path. The short stub in `squad.agent.md` tells you to load this skill; the full two-phase protocol lives here.
>
> **⚠<U+FE0F> Eager-execution exception:** Init Mode is the ONE exception to the eager-execution / parallel-fan-out doctrine. Phase 1 MUST end with a user confirmation before any file is created. Do not bypass.
## Phase 1: Propose the Team
No team exists yet. **Propose one — but DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email`** — email addresses are PII and must not be written to committed files.
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see the canonical Casting reference at `.squad/templates/casting-reference.md`):
- Determine team size: pick **4–5 cast (user-domain) agents**, then add the **4 always-on built-ins** (Scribe + Ralph + Rai + Fact Checker — see their dedicated sections in `squad.agent.md`). A typical fresh squad has **8–9 total roster entries**, not 4–5.
- Determine assignment shape from the user's project description.
- Derive resonance signals from the session and repo context.coordinator-response-mode
When your agent uses it
Selecting WHO handles work is the Routing table; selecting HOW they handle it (Direct, Lightweight, Standard, Full) is Response Mode. This skill contains the complete decision table, exemplar prompts for each mode, the Lightweight spawn template, and the upgrade rules. Squad coordinator loads this on demand once routing has identified the agent — to pick the right ceremony level for the task.
What it loads into context
Frontmatter
139 tokensAlways available to the agent
SKILL.md
1,172 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Response Mode Selection
- Direct Mode exemplars
- Lightweight Mode exemplars
- Standard Mode exemplars
- Full Mode exemplars
- Mode upgrade rules
- Lightweight Spawn Template
Files
SKILL.mdcore5,018 bytes
Read SKILL.md
> **Load this skill when:** you have routed work to an agent and need to pick the response mode (Direct / Lightweight / Standard / Full). The 1-line stub in `squad.agent.md` is for awareness; this skill is the full decision table + templates.
## Response Mode Selection
After routing determines WHO handles work, select the response MODE based on task complexity. **Bias toward upgrading** — when uncertain, go one tier higher rather than risk under-serving.
| Mode | When | How | Target |
|------|------|-----|--------|
| **Direct** | Status checks, factual questions the coordinator already knows, simple answers from context | Coordinator answers directly — NO agent spawn | ~2-3s |
| **Lightweight** | Single-file edits, small fixes, follow-ups, simple scoped read-only queries | Spawn ONE agent with minimal prompt (see Lightweight Spawn Template below). Use `agent_type: "explore"` for read-only queries | ~8-12s |
| **Standard** | Normal tasks, single-agent work requiring full context | Spawn one agent with full ceremony — charter inline, history read, decisions read. This is the current default | ~25-35s |
| **Full** | Multi-agent work, complex tasks touching 3+ concerns, "Team" requests | Parallel fan-out, full ceremony, Scribe included | ~40-60s |
## Direct Mode exemplarscoordinator-source-of-truth
When your agent uses it
The complete file-by-file source-of-truth hierarchy for Squad: which files are authoritative, which are derived/append-only, who may write each one, who may read each one, and the precedence rules when they conflict. Squad coordinator loads this on demand when it needs to resolve a write conflict, decide where a piece of state belongs, or answer a 'who owns this file' question.
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
1,380 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- State backend note
- File hierarchy
- Rules
Files
SKILL.mdcore5,941 bytesSKILL.mdcore5,941 bytes
Read SKILL.md
> **Load this skill when:** the coordinator (or any agent) needs to resolve a "where does this state belong?" or "who is allowed to write this file?" question — e.g., when about to write `.squad/decisions.md`, when reviewing whether an agent broke the append-only rule, when answering a user question about Squad's file layout, or when adjudicating a conflict between two files. The short summary in `squad.agent.md` is for routing; this skill is the full reference.
## State backend note
Files below marked as **"Derived / append-only"** are **mutable state** — agents access them with runtime state tools (`squad_state_read`, `squad_state_write`, `squad_state_append`, `squad_state_create_if_absent`, `squad_state_delete`, `squad_state_list`). The runtime decides whether the configured backend stores them on disk, git-native state, or an external provider. Files marked as **"Authoritative"** are **static config** and always live on disk regardless of backend.
## File hierarchy
| File | Status | Who May Write | Who May Read |
|------|--------|---------------|--------------|
| `.github/agents/squad.agent.md` | **Authoritative governance.** All roles, handoffs, gates, and enforcement rules. | Repo maintainer (human) | Squad (Coordinator) |
| `.squad/decisions.md` | **Authoritative decision ledger.** Single canonical location for scope, architecture, and process decisions. | Squad (Coordinator) — append only | All agents |
| `.squad/team.md` | **Authoritative roster.** Current team composition. | Squad (Coordinator) | All agents |
| `.squad/routing.md` | **Authoritative routing.** Work assignment rules. | Squad (Coordinator) | Squad (Coordinator) |cross-machine-coordination
When your agent uses it
Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002).
What it loads into context
Frontmatter
58 tokensAlways available to the agent
SKILL.md
3,211 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Usage
- File Formats
- Security Model
- Configuration
- Examples
- Error Handling
- Monitoring & Debugging
- Integration with Ralph Watch
- Migration from Manual Handoff
- Future Enhancements
- Questions?
Files
SKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytes
Read SKILL.md
# Skill: Cross-Machine Coordination Pattern
**Skill ID:** `cross-machine-coordination`
**Owner:** Ralph (Work Monitor)
**Squad Integration:** All agents
**Status:** Design proposal — NOT implemented
> ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
> the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
> and `squad watch` do not poll task files, run whitelisted commands, or
> write result files. Everything below describes a proposed design, not
> current behavior. Do not tell users it works, and do not create task files
> expecting another machine to pick them up.
>cross-squad
When your agent uses it
Coordinating work across multiple Squad instances — discovery, delegation, and disambiguation when the user says 'squad' (the product) vs casual English 'group of agents'.
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
2,402 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Disambiguation: "squad" vs ad-hoc agents
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytes
Read SKILL.md
## Context
> **Read this FIRST any time the user says "squad" as a thing to spawn, delegate to, address, or fan out to** — e.g., *"spawn two squads of designers and devs"*, *"ask the other squad"*, *"delegate to a squad"*. In Squad-PRODUCT vocabulary, "squad" is a **peer** (an independent installation with its own `.squad/`, `team.md`, MCP server, and agents) — NOT a generic English synonym for "team" or "group". Do not fan out raw `task` agents inside your own coordinator context when the user means "another squad". Use the discovery and communication patterns below (and the companion `cross-squad-communication` skill for the actual protocols).
When an organization runs multiple Squad instances (e.g., platform-squad, frontend-squad, data-squad), those squads need to discover each other, share context, and hand off work across repository boundaries. This skill teaches agents how to coordinate across squads without creating tight coupling.
> **Companion skill — for protocol details:** `cross-squad-communication/SKILL.md` covers the four communication patterns (synchronous CLI, read-only knowledge query, git-based async, and GitHub-issue-based delegation) once a peer squad is discovered via the registry below. This skill answers "who?" — the companion answers "how?".
Cross-squad orchestration applies when:
- A task requires capabilities owned by another squad
- An architectural decision affects multiple squads
- A feature spans multiple repositories with different squads
- A squad needs to request infrastructure, tooling, or support from another squad
cross-squad-communication
When your agent uses it
Protocol for sending queries, delegating tasks, and sharing context between independent Squad instances across different repositories
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
4,739 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Universal rule: every `copilot` spawn into a peer squad MUST pass `--agent squad`
- Discovery Protocol
- Platform Compatibility Matrix
- Examples
- Anti-Patterns
- Validation Status
Files
SKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytes
Read SKILL.md
## Context
When multiple repositories each have their own Squad (AI team), they need to exchange information: knowledge queries, PR reviews, task delegation, and dependency analysis. Each squad has its own agents, MCP tools, and issue tracker — there is no shared runtime.
> **Companion skill — read first:** `cross-squad/SKILL.md` covers **discovery** of peer squads via `squad registry add/list/remove`. This skill picks up after a peer is known and covers the **communication protocols** themselves — the four numbered patterns below: Pattern 0 (synchronous CLI), Pattern 1 (read-only knowledge query), Pattern 2 (git-based async), and Pattern 3 (GitHub-issue-based delegation). A separate non-numbered appendix (Cross-Repo Dependency Scan) is provided as a related analysis tool, not a communication pattern. The two skills are designed to be used together.
**When this skill applies:**
- A squad agent needs information from another squad-enabled repo
- A task needs to be delegated to another squad
- Cross-repo dependency analysis is needed
- PR review requests span repo boundaries
**Key constraint:** Each squad has its own runtime, MCP tools, and issue tracker. Cross-squad communication can be **synchronous** (via CLI session targeting the other repo) or **asynchronous** (file-based or issue-based). The coordinator decides which approach fits.
distributed-mesh
When your agent uses it
How to coordinate with squads on different machines using git as transport
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
3,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Patterns
- Examples
- AGENT WORKFLOW (Deterministic Setup)
- Anti-Patterns
Scripts executed separately
sync-mesh.ps1sync-mesh.sh
Files
SKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesmesh.json.exampleasset762 bytessync-mesh.ps1runs3,861 bytessync-mesh.shruns3,395 bytes
Read SKILL.md
## SCOPE
**✅ THIS SKILL PRODUCES (exactly these, nothing more):**
1. **`mesh.json`** — Generated from user answers about zones and squads (which squads participate, what zone each is in, paths/URLs for each), using `mesh.json.example` in this skill's directory as the schema template
2. **`sync-mesh.sh` and `sync-mesh.ps1`** — Copied from this skill's directory into the project root (these are bundled resources, NOT generated code)
3. **Zone 2 state repo initialization** (if applicable) — If the user specified a Zone 2 shared state repo, run `sync-mesh.sh --init` to scaffold the state repo structure
4. **A decision entry** in `.squad/decisions/inbox/` documenting the mesh configuration for team awareness
**❌ THIS SKILL DOES NOT PRODUCE:**
- **No application code** — No validators, libraries, or modules of any kind
- **No test files** — No test suites, test cases, or test scaffolding
- **No GENERATING sync scripts** — They are bundled with this skill as pre-built resources. COPY them, don't generate them.docs-standards
When your agent uses it
Microsoft Style Guide + Squad-specific documentation patterns
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
524 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore2,270 bytesSKILL.mdcore2,270 bytes
Read SKILL.md
## Context
Squad documentation follows the Microsoft Style Guide with Squad-specific conventions. Consistency across docs builds trust and improves discoverability.
## Patterns
### Microsoft Style Guide Rules
- **Sentence-case headings:** "Getting started" not "Getting Started"
- **Active voice:** "Run the command" not "The command should be run"
- **Second person:** "You can configure..." not "Users can configure..."
- **Present tense:** "The system routes..." not "The system will route..."
- **No ampersands in prose:** "and" not "&" (except in code, brand names, or UI elements)
### Squad Formatting Patternse2e-template-testing
When your agent uses it
End-to-end validation of coordinator and agent template changes
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
6,164 tokensWhen the skill is activated
Reference documents
5,121 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- When To Use
- Prerequisites
- Workflow
- Progress Reporting
- Test Matrix Template
- Tips
- Fast-Fail Rules
- PII Protection — Mandatory
- Anti-Patterns
- Sandbox / Permission Notes
- Confidence
Files
SKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytes../../../CONTRIBUTING.mddoc21,149 bytes
Read SKILL.md
## Context
Squad's coordinator prompt (`squad.agent.md`) and agent charters (e.g.
`scribe-charter.md`) are shipped as templates in `.squad-templates/`. Changes to
these files affect how every squad session behaves — but unit tests can't catch
prompt-level regressions because the prompts are interpreted by an LLM at
runtime.
This skill describes how to validate template changes end-to-end by running real
squad sessions against a locally-built CLI that includes your modified templates.
## When To Use
- You changed `.squad-templates/squad.agent.md` (coordinator prompt)economy-mode
When your agent uses it
Shifts Layer 3 model selection to cost-optimized alternatives when economy mode is active.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,396 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Activation Methods
- Economy Model Selection Table
- AGENT WORKFLOW
- Config Schema
- Anti-Patterns
Files
SKILL.mdcore5,547 bytes
Read SKILL.md
## SCOPE
✅ THIS SKILL PRODUCES:
- A modified Layer 3 model selection table applied when economy mode is active
- `economyMode: true` written to `.squad/config.json` when activated persistently
- Spawn acknowledgments with `💰` indicator when economy mode is active
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Cost reports or billing artifacts
- Changes to Layer 0, Layer 1, or Layer 2 resolution (user intent always wins)
## Context
error-recovery
When your agent uses it
Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
785 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Retry with Backoff
- 2. Fallback Alternatives
- 3. Diagnose-and-Fix
- 4. Escalate with Context
- 5. Graceful Degradation
- Applying These Patterns
Files
SKILL.mdcore3,466 bytesSKILL.mdcore3,466 bytes
Read SKILL.md
# Error Recovery Patterns
Standard recovery patterns for all squad agents. When something fails, **adapt** — don't just report the failure.
---
## 1. Retry with Backoff
**When:** Transient failures — API timeouts, rate limits, network errors, temporary service unavailability.
**Pattern:**
1. Wait briefly, then retry (start at 2s, double each attempt)
2. Maximum 3 retries before escalating
3. Log each attempt with the error receivedexternal-comms
When your agent uses it
Workflow for scanning, drafting, and presenting community responses with a human review gate
What it loads into context
Frontmatter
228 tokensAlways available to the agent
SKILL.md
2,993 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore12,847 bytesSKILL.mdcore12,847 bytes
Read SKILL.md
## Context
Phase 1 is **draft-only mode**.
- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.
- **Human review gate is mandatory** — the communications owner never posts autonomously.
- Every action is logged to `.squad/comms/audit/`.
- This workflow is triggered manually only ("check community") — no automated activation in Phase 1.
## Patterns
### 1. Scan
Find unanswered community items with GitHub MCP tools first, or `gh issue list` / `gh api` as fallback for issues and discussions.fact-checking
When your agent uses it
Review and validate claims using counter-hypothesis testing. Use when verifying technical content, checking references, validating API endpoints, or performing quality assurance on deliverables.
What it loads into context
Frontmatter
97 tokensAlways available to the agent
SKILL.md
469 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore2,046 bytes
Read SKILL.md
# Skill: Fact Checking
## Context
Codifies the challenger agent review output format and methodology so any agent performing fact-checking or review produces consistent, structured output.
## Pattern
### Review Methodology
For every claim or deliverable under review:
1. Ask: "What evidence supports this? What would disprove it?"
2. Generate counter-hypotheses and test them against available data
3. Verify URLs, package names, API endpoints, and external references actually exist
4. Flag confidence levels: ✅ Verified, ⚠<U+FE0F> Unverified, ❌ Contradictedgh-auth-isolation
When your agent uses it
Safely manage multiple GitHub identities (EMU + personal) in agent workflows
What it loads into context
Frontmatter
100 tokensAlways available to the agent
SKILL.md
1,537 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytes
Read SKILL.md
## Context
Many developers use GitHub through an Enterprise Managed User (EMU) account at work while maintaining a personal GitHub account for open-source contributions. AI agents spawned by Squad inherit the shell's default `gh` authentication — which is usually the EMU account. This causes failures when agents try to push to personal repos, create PRs on forks, or interact with resources outside the enterprise org.
This skill teaches agents how to detect the active identity, switch contexts safely, and avoid mixing credentials across operations.
## Patterns
### Detect Current Identity
Before any GitHub operation, check which account is active:
```bash
gh auth statusgh-aw-enlistment
When your agent uses it
Enlist a repository into Squad by installing the supported GitHub Agentic Workflows (gh-aw) bootstrap — with strict compilation, an explicit safe-update allowlist, and a human-reviewed bootstrap PR. Never auto-merges.
What it loads into context
Frontmatter
308 tokensAlways available to the agent
SKILL.md
7,239 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytes
Read SKILL.md
## Context
Use this skill when a user wants to **enlist a repository into Squad** through
**GitHub Agentic Workflows (`gh aw`)** — e.g. "set up Squad agentic workflows",
"enlist this repo in Squad", or "install Squad gh-aw workflows". The end state is
the `/squad` slash command being live on the repo, delivered as a **human-reviewed
bootstrap pull request** — never an auto-merge.
This skill **operationalizes** the supported bootstrap path. It does not summarize
it: each step below is a gate with explicit success evidence and **STOP conditions**.
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this
skill ever disagree, the guide wins — re-read it before proceeding.
git-workflow
When your agent uses it
Squad branching model: dev-first workflow with insiders preview channel
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,581 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Branch Naming Convention
- Workflow for Issue Work
- Parallel Multi-Issue Work (Worktrees)
- Multi-Repo Downstream Scenarios
- Anti-Patterns
- Promotion Pipeline
Files
SKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytes
Read SKILL.md
## Context
Squad uses a three-branch model. **All feature work starts from `dev`, not `main`.**
| Branch | Purpose | Publishes |
|--------|---------|-----------|
| `main` | Released, tagged, in-npm code only | `npm publish` on tag |
| `dev` | Integration branch — all feature work lands here | `npm publish --tag preview` on merge |
| `insiders` | Early-access channel — synced from dev | `npm publish --tag insiders` on sync |
## Branch Naming Convention
Issue branches MUST use: `squad/{issue-number}-{kebab-case-slug}`
github-multi-account
When your agent uses it
Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
What it loads into context
Frontmatter
80 tokensAlways available to the agent
SKILL.md
1,038 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Activate
- What to Do (as the AI agent)
- After Setup — Usage Rules
- Repo-Specific Account Binding
- For Squad Agents
Files
SKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytes
Read SKILL.md
# GitHub Multi-Account — AI-Driven Setup
## When to Activate
When the user has multiple GitHub accounts (check with `gh auth status`). If you see 2+ accounts listed, this skill applies.
## What to Do (as the AI agent)
### Step 1: Detect accounts
Run: `gh auth status`
Look for multiple accounts. Note which usernames are listed.
### Step 2: Ask the user
Ask: "I see you have multiple GitHub accounts: {list them}. Which one is your personal account and which is your work/EMU account?"
history-hygiene
When your agent uses it
Record reconciled final outcomes to history.md, not intermediate requests or obsolete claims
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
545 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Patterns
- Examples
- Stop conditions
- Anti-Patterns
Files
SKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytes
Read SKILL.md
## Context
History files are cold-start context, not the source of truth for live behavior. Stale or incorrect
entries poison decision-making downstream. A stale-history incident proved that an old history claim
can cause future spawns to repeat a reversed decision. Reconcile an operational claim with current
implementation, workflows, tests, documentation, skills, and applicable decisions before acting on it.
## Scope
- Activate when recording, correcting, or promoting finalized operational history.
- Allowed targets: the explicitly selected history file and a focused reusable skill when the rule
applies beyond one role.
- Exclusions: intermediate requests, stale assignments, roster metadata, and unrelated cleanup.
humanizer
When your agent uses it
Tone enforcement patterns for external-facing community responses
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,004 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,330 bytes
Read SKILL.md
## Context
Use this skill whenever the communications owner drafts external-facing responses for issues or discussions.
- Tone must be warm, helpful, and human-sounding — never robotic or corporate.
- The repository's communication policy applies everywhere: **Humanized tone is mandatory**.
- This applies to **all external-facing content** drafted by the communications owner in Phase 1 issues/discussions workflows.
## Patterns
1. **Warm opening** — Start with acknowledgment ("Thanks for reporting this", "Great question!")
2. **Active voice** — "We're looking into this" not "This is being investigated"
3. **Second person** — Address the person directly ("you" not "the user")
4. **Conversational connectors** — "That said...", "Here's what we found...", "Quick note:"init-mode
When your agent uses it
Team initialization flow (Phase 1 proposal + Phase 2 creation)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
1,749 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,100 bytesSKILL.mdcore7,100 bytes
Read SKILL.md
## Context
Init Mode activates when `.squad/team.md` does not exist, or exists but has zero roster entries under `## Members`. The coordinator proposes a team (Phase 1), waits for user confirmation, then creates the team structure (Phase 2).
## Patterns
### Phase 1: Propose the Team
No team exists yet. Propose one — but **DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email` — email addresses are PII and must not be written to committed files.**
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see that section):
- Determine team size (typically 4–5 + Scribe).iterative-retrieval
When your agent uses it
Max-3-cycle protocol for agent sub-tasks with WHY context and coordinator validation. Use when spawning sub-agents to complete scoped work.
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,385 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Spawn Prompt Template
- 3-Cycle Protocol
- Coordinator Validation Checklist
- When to Escalate vs Retry
- Issue Dedup Check (Mandatory)
- Mandatory Output Requirement (Research-Then-Execute)
- Anti-Patterns
- Examples
Files
SKILL.mdcore6,341 bytesSKILL.mdcore6,341 bytes
Read SKILL.md
# Iterative Retrieval Skill
Squad agents frequently spawn sub-agents to complete scoped work. Without structure, these
handoffs become vague, cycles multiply, and outputs land without being checked. The
**Iterative Retrieval Pattern** caps cycles at 3, mandates WHY context in every spawn, and
requires the coordinator to validate agent output before closing an issue.
---
## Spawn Prompt Template
Every agent spawn must include the following four sections. Copy and fill in the template:
```model-selection
When your agent uses it
Determines which LLM model to use for each agent spawn
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- 5-Layer Model Resolution Hierarchy
- AGENT WORKFLOW
- Config Schema
- Fallback Chains
Files
SKILL.mdcore8,082 bytes
Read SKILL.md
# Model Selection
> Determines which LLM model to use for each agent spawn.
## SCOPE
✅ THIS SKILL PRODUCES:
- A resolved `model` parameter for every `task` tool call
- Persistent model preferences in `.squad/config.json`
- Spawn acknowledgments that include the resolved model
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Model performance benchmarksnap
When your agent uses it
Context hygiene — compress, prune, archive .squad/ state
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
208 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What It Does
- When To Use
- Invocation
- Confidence
Files
SKILL.mdcore792 bytes
Read SKILL.md
# Skill: nap
> Context hygiene — compress, prune, archive .squad/ state
## What It Does
Reclaims context window budget by compressing agent histories, pruning old logs,
archiving stale decisions, and cleaning orphaned inbox files.
## When To Use
- Before heavy fan-out work (many agents will spawn)
- When history.md files exceed 15KB
- When .squad/ total size exceeds 1MBnotification-routing
When your agent uses it
Route agent notifications to specific channels by type — prevent alert fatigue from single-channel flooding
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
730 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
- Distributed Systems Pattern
Files
SKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytes
Read SKILL.md
## Context
When a Squad grows beyond a few agents, notifications flood a single channel — failure alerts drown in daily
briefings, tech news buries security findings, and everything gets ignored. This is the pub-sub problem:
a single message queue for everything is a recipe for missed alerts.
The fix is **topic-based routing**: agents tag notifications with a channel type, and a routing function
sends them to the appropriate destination.
**Trigger symptoms:**
- Important alerts missed because they're buried in routine notifications
- Team members turning off notifications entirely (signal overwhelm)
- Onboarding friction: "where do I look for X?"
pr-review-response
When your agent uses it
Teaches agents to reply to PR review comment threads after fixing issues, making resolutions traceable
What it loads into context
Frontmatter
154 tokensAlways available to the agent
SKILL.md
2,574 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- SCOPE
- Patterns
- AGENT WORKFLOW (Summary)
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytes
Read SKILL.md
## Context
When an agent fixes code in response to PR review comments (from Copilot, a human reviewer, or any GitHub reviewer), the fix alone is not enough. The reviewer needs to see — on the PR thread itself — which comments were addressed and how. Without replies, comments stay visually unresolved, reviewers must re-read the entire diff to verify fixes, and there's no traceable link between feedback and resolution.
Use this skill whenever:
- You are fixing code based on PR review feedback
- You are addressing Copilot review suggestions
- You are responding to reviewer-requested changes on a PR
- A squad member hands you review comments to resolve
## SCOPE
✅ THIS SKILL PRODUCES:
- Reply comments on each review thread explaining the fixpr-screenshots
When your agent uses it
Capture Playwright screenshots and embed them in GitHub PR descriptions
What it loads into context
Frontmatter
62 tokensAlways available to the agent
SKILL.md
1,058 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../scripts/capture-pr-screenshots.mjs
Files
SKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytes../../../scripts/capture-pr-screenshots.mjsruns1,855 bytes
Read SKILL.md
## Context
When a PR includes visual changes (docs sites, UI components, generated pages), reviewers
need to see what the PR delivers without checking out the branch. Screenshots belong in
the **PR description body**, not as committed files and not as text descriptions.
Use this skill whenever:
- A PR touches docs site pages (Astro, Starlight, etc.)
- A PR adds or changes UI components
- A PR generates visual artifacts (TypeDoc, Storybook, diagrams)
- Playwright tests already capture screenshots as part of testing
## Patterns
project-conventions
When your agent uses it
Core conventions and patterns for this codebase
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,555 bytes
Read SKILL.md
## Context
> **This is a starter template.** Replace the placeholder patterns below with your actual project conventions. Skills train agents on codebase-specific practices — accurate documentation here improves agent output quality.
## Patterns
### [Pattern Name]
Describe a key convention or practice used in this codebase. Be specific about what to do and why.
### Error Handling
<!-- Example: How does your project handle errors? -->
<!-- - Use try/catch with specific error types? -->ralph-two-pass-scan
When your agent uses it
Cuts GitHub API calls by separating lightweight list scanning from full hydration
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
345 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore1,336 bytes
Read SKILL.md
# Skill: Ralph — Two-Pass Issue Scanning
**Confidence:** high
**Domain:** work-monitoring
**Last validated:** 2026-03-24
## Context
Cuts GitHub API calls from N+1 to ~7 per round (~72% reduction) by separating list scanning from full hydration.
Addresses the scanning inefficiency described in issue #596.
## Pattern
### Pass 1 — Lightweight Scan
```reflect
When your agent uses it
Learning capture system that extracts HIGH/MED/LOW confidence patterns from conversations to prevent repeating mistakes. Use after user corrections ("no", "wrong"), praise ("perfect", "exactly"), or when discovering edge cases. Complements .squad/agents/{agent}/history.md and .squad/decisions.md.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Integration with Squad Architecture
- Triggers
- Process
- Usage Examples
- When to Use
- See Also
Files
SKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytes
Read SKILL.md
# Reflect Skill
**Critical learning capture system** for Squad. Prevents repeating mistakes and preserves successful patterns across sessions.
Analyze conversations and propose improvements to squad knowledge based on what worked, what didn't, and edge cases discovered. **Every correction is a learning opportunity.**
---
## Integration with Squad Architecture
**Reflect complements existing Squad knowledge systems:**
1. **Governed runtime memory** — Durable observations and decisions persisted through configured state tools
2. **Tracked governing source** — Coordinator prompts, skills, and other behavior-defining files changed through focused pull requestsrelease-process
When your agent uses it
Prepare, publish, recover, and verify Squad insider, preview, and stable releases
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
2,578 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Non-negotiable rules
- Required credentials
- Human release-trigger boundary
- Prepare a release
- Publish a preview release
- Publish an insider release
- Publish a stable release
- Verify publication
- Recovery
- After a stable release
Files
SKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytes
Read SKILL.md
# Release Process
This is the canonical Squad release runbook. Squad has three release channels
and two long-lived branches:
| Source | Version | GitHub | npm | Standalone | Homebrew/WinGet |
|--------|---------|--------|-----|------------|-----------------|
| Insider dispatch from `dev` | Generated `X.Y.Z-insider.N` | Prerelease | `insider` | Yes | Yes |
| Release dispatch from `dev` | `X.Y.Z-preview.N` | Prerelease | `preview` | Yes | Yes |
| Promotion PR merged to `main` (or legacy workflow push) | `X.Y.Z` | Stable/latest | `latest` | Yes | Yes |
The repository does not use a staging `preview` branch. Preview is a release
channel, not a branch.
reskill
When your agent uses it
Extract durable operating knowledge before slimming or replacing team charters
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,223 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Process
- Patterns
- Anti-Patterns
Files
SKILL.mdcore5,760 bytesSKILL.mdcore5,760 bytes
Read SKILL.md
## Context
When the coordinator hears "team, reskill" (or similar: "optimize context", "slim down charters"), trigger a team-wide optimization pass. The goal: reduce per-agent context consumption by extracting shared patterns from charters and histories into reusable skills.
This is a periodic maintenance activity. Run whenever charter/history bloat is suspected or a
team will be re-cast. Charters and histories are evidence, not the authority for current product
behavior: reconcile candidates against current source, workflows, tests, documentation, skills,
and decisions before preserving them.
## Process
### Step 1: Audit
Read all active agent charters and histories, then inventory existing skills. Measure byte sizes.
Identify:When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
1,352 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Purpose
- Core Function: Test-RetroOverdue
- Coordinator Integration
- Action Item Enforcement
- Cadence Enforcement
- Skill Metadata
Files
SKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytes
Read SKILL.md
# Skill: Retro Enforcement
## Purpose
Ensure retrospectives happen on schedule and that their action items are tracked in GitHub Issues — not markdown checklists.
This skill addresses a specific, measured failure mode: **0% completion rate on markdown retro action items across 6 consecutive retrospectives**. GitHub Issues have an 85%+ completion rate in the same squad. The format was the problem, not the people.
## Core Function: Test-RetroOverdue
```powershell
function Test-RetroOverdue {
param(
[string]$LogDir = ".squad/log",
[int]$WindowDays = 7,reviewer-protocol
When your agent uses it
Reviewer rejection workflow and strict lockout semantics
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
1,009 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,658 bytes
Read SKILL.md
## Context
When a team member has a **Reviewer** role (e.g., Tester, Code Reviewer, Lead), they may approve or reject work from other agents. On rejection, the coordinator enforces strict lockout rules to ensure the original author does NOT self-revise. This prevents defensive feedback loops and ensures independent review.
## Patterns
### Reviewer Rejection Protocol
Quality and test reviewers may block a PR or merge on quality, coverage, or regression grounds. Treat that block as a rejection that enters the protocol below.
When a team member has a **Reviewer** role:
- Reviewers may **approve** or **reject** work from other agents.
- On **rejection**, the Reviewer may choose ONE of:secret-handling
When your agent uses it
Never read .env files or write secrets to .squad/ committed files
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
1,994 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,524 bytesSKILL.mdcore7,524 bytes
Read SKILL.md
## Context
Spawned agents have read access to the entire repository, including `.env` files containing live credentials. If an agent reads secrets and writes them to `.squad/` files (decisions, logs, history), Scribe auto-commits them to git, exposing them in remote history. This skill codifies absolute prohibitions and safe alternatives.
## Patterns
### Prohibited File Reads
**NEVER read these files:**
- `.env` (production secrets)
- `.env.local` (local dev secrets)
- `.env.production` (production environment)
- `.env.development` (development environment)
- `.env.staging` (staging environment)session-recovery
When your agent uses it
Find and resume interrupted Copilot CLI sessions using session_store queries
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,215 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytes
Read SKILL.md
## Context
Squad agents run in Copilot CLI sessions that can be interrupted — terminal crashes, network drops, machine restarts, or accidental window closes. When this happens, in-progress work may be left in a partially-completed state: branches with uncommitted changes, issues marked in-progress with no active agent, or checkpoints that were never finalized.
Copilot CLI stores session history in a SQLite database called `session_store` (read-only, accessed via the `sql` tool with `database: "session_store"`). This skill teaches agents how to query that store to detect interrupted sessions and resume work.
## Patterns
### 1. Find Recent Sessions
Query the `sessions` table filtered by time window. Include the last checkpoint to understand where the session stopped:
```sql
SELECTsquad
When your agent uses it
Squad's command catalog and interactive menu. Invoke via /squad (slash command) or natural language ("squad commands", "what can squad do", "show me squad options"). Presents categorized operations (Install & Upgrade, Team Management, Issues & PRs, Plugins & Skills, Model & Cost, Sessions & State) as an interactive picker. Routes to the right squad CLI command or the Squad coordinator agent.
What it loads into context
Frontmatter
103 tokensAlways available to the agent
SKILL.md
2,680 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,874 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill (via `/squad` slash command, "squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-commands
When your agent uses it
Categorized catalog of common Squad operations. Coordinator reads this file and presents it as an interactive menu when the user asks for available commands or help.
What it loads into context
Frontmatter
82 tokensAlways available to the agent
SKILL.md
2,652 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,748 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill ("squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-conventions
When your agent uses it
Core conventions and patterns used in the Squad codebase
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
824 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore3,440 bytesSKILL.mdcore3,440 bytes
Read SKILL.md
## Context
These conventions apply to all work on the Squad CLI tool (`create-squad`). Squad is a zero-dependency Node.js package that adds AI agent teams to any project. Understanding these patterns is essential before modifying any Squad source code.
## Patterns
### Zero Dependencies
Squad has zero runtime dependencies. Everything uses Node.js built-ins (`fs`, `path`, `os`, `child_process`). Do not add packages to `dependencies` in `package.json`. This is a hard constraint, not a preference.
### Node.js Built-in Test Runner
Tests use `node:test` and `node:assert/strict` — no test frameworks. Run with `npm test`. Test files live in `test/`. The test command is `node --test test/`.
### Error Handling — `fatal()` Pattern
All user-facing errors use the `fatal(msg)` function which prints a red `✗` prefix and exits with code 1. Never throw unhandled exceptions or print raw stack traces. The global `uncaughtException` handler calls `fatal()` as a safety net.
squad-help
When your agent uses it
How to actually use Squad — Squad is a custom Copilot agent (invoked via the task tool with agent_type='Squad'), not a skill. This file explains the right invocation paths for setting up a team, listing squad commands, and initializing Squad in a new project.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,043 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Squad is a custom agent, not a skill
- How to actually use Squad
- What NOT to do
- How this skill was discovered
- See also
Files
SKILL.mdcore4,198 bytesSKILL.mdcore4,198 bytes
Read SKILL.md
# Skill: squad-help
> **Quick reference.** If you're reading this because a user said "use squad" or "squad" or "set up a squad", you're in the right place — read on for the correct invocation paths.
---
## Squad is a custom agent, not a skill
The Squad framework registers a **custom Copilot CLI agent** at `.github/agents/squad.agent.md`. The agent is named **`Squad`** and its description is *"Your AI team. Describe what you're building, get a team of specialists that live in your repo."*
Copilot CLI agents and skills are different things:
| Thing | How to invoke | Example |
|---|---|---|squad-version-check
When your agent uses it
Internals of how @bradygaster/squad-cli stamps its version, how `squad upgrade` works (what it preserves vs overwrites), and how to probe the npm registry for the latest version from a coordinator prompt.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,702 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What This Skill Covers
- Package & Registry Facts
- Version Stamping Mechanism
- `squad upgrade` Behavior
- `.squad/config.json` — What It Holds
- Version-Check Probe (npm Registry)
- Upstream OS-Specific Cache
- Repo-Local Cache Convention: `.squad/.cache/version-check.json`
- Key File Paths (installed CLI)
Files
SKILL.mdcore6,296 bytes
Read SKILL.md
# SKILL: Squad CLI Internals — Version Stamping & Upgrade Mechanics
**Confidence:** medium
**Discovered by:** repository investigation
**Date:** 2026-05-26
**Validated in:** Issue #1173 recon (bradygaster/squad)
---
## What This Skill Covers
Reusable knowledge about how `@bradygaster/squad-cli` stamps its version into `squad.agent.md`, how `squad upgrade` works, what it preserves vs. overwrites, and how to probe the npm registry for the latest version from a coordinator prompt.
---test-discipline
When your agent uses it
Design assertions that prove the actual contract and fail on realistic regressions
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytes
Read SKILL.md
## Context
Use this skill whenever a change affects an API, public interface, workflow contract, generated
artifact, or user-observable behavior. A passing status-only test is not evidence that the
intended artifact or contract is correct.
## Patterns
- Update tests in the same change as an API, signature, export, or public-interface change.
- Make assertions name and inspect the offending artifact, output, or configuration field—not just
an exit code or broad success status.
- Mutation-test critical gates by changing the real source or emitted artifact and proving the
assertion turns red.
- Exercise every caller path for a changed parser, normalizer, or helper; isolated unit coveragetiered-memory
When your agent uses it
Three-tier agent memory model (hot/cold/wiki) for context reduction per spawn
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
1,775 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Memory Tiers
- When to Load Each Tier
- Spawn Template Pattern
- Integration with Scribe Agent (design — not yet implemented)
- Implementation Checklist (tracked in #1264)
- References
- Spawn Template
- Task
- WHY
- Success Criteria
- Memory Context
- Escalation
- Notes
Files
SKILL.mdcore7,506 bytes
Read SKILL.md
# Skill: Tiered Agent Memory
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, and spawn-template tier-aware reads are tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264). Until those land, agents continue to load full `history.md` + `decisions.md` on every spawn.
## Overview
Squad agents today load their full context history on every spawn, which grows unboundedly across sessions. The Tiered Agent Memory model proposes a three-tier separation so agents only load the bytes that are actually relevant to the current task, with older context kept available on demand.
---
## Memory Tiers
### 🔥 Hot Tier — Current Session Context
- **Size target:** keep small (~2–4KB typical)versioning-policy
When your agent uses it
SemVer rules for Squad's stable, preview, insider, and local package versions
What it loads into context
Frontmatter
60 tokensAlways available to the agent
SKILL.md
1,153 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- 1. Supported versions
- 2. Package versions stay in lockstep
- 3. Prerelease workspace dependency rule
- 4. Local build versions are ephemeral
- 5. Release lifecycle
- 6. Ownership
- 7. CI enforcement
- Quick reference
Scripts executed separately
../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytes../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
Squad publishes `@bradygaster/squad-sdk` and
`@bradygaster/squad-cli` from one npm workspace. Their versions and the root
version move together. A mismatched prerelease dependency can make npm silently
resolve an older registry SDK instead of the local workspace.
## 1. Supported versions
| Use | Version | Committed branch |
|---|---|---|
| Stable release | `MAJOR.MINOR.PATCH` | `main`, and briefly `dev` during promotion |
| Preview release | `MAJOR.MINOR.PATCH-preview.N` | `dev` |
| Insider snapshot | `MAJOR.MINOR.PATCH-insider.N` | Generated by the insider workflow |windows-compatibility
When your agent uses it
Cross-platform path handling and command patterns
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
1,572 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../scripts/fix-crlf-worktree.mjs
Files
SKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytes../../../scripts/fix-crlf-worktree.mjsruns5,444 bytes
Read SKILL.md
## Context
Squad runs on Windows, macOS, and Linux. Several bugs have been traced to platform-specific assumptions: ISO timestamps with colons (illegal on Windows), `git -C` with Windows paths (unreliable), forward-slash paths in Node.js on Windows.
## Patterns
### Filenames & Timestamps
- **Never use colons in filenames:** ISO 8601 format `2026-03-15T05:30:00Z` is illegal on Windows
- **Use `safeTimestamp()` utility:** Replaces colons with hyphens → `2026-03-15T05-30-00Z`
- **Centralize formatting:** Don't inline `.toISOString().replace(/:/g, '-')` — use the utility
### Git Commands
- **Never use `git -C {path}`:** Unreliable with Windows paths (backslashes, spaces, drive letters)
- **Always `cd` first:** Change directory, then run git commandsmodel-selection
When your agent uses it
Resolve explicit and persistent model preferences for Squad spawns without pinning stale defaults
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
434 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Resolution Order
- Persistent Preferences
- Spawn Behavior
- Stop Conditions
Files
SKILL.mdcore2,251 bytes
Read SKILL.md
# Model Selection
Use this skill when the user requests a model, the team configuration contains a saved model
preference, or a specialist charter explicitly requires one.
## Resolution Order
Apply the first available value:
1. Explicit model requested for the current spawn.
2. Current-session model directive from the user.
3. `.squad/config.json` `agentModelOverrides.{specialist-id}`.
4. `.squad/config.json` `defaultModel`.
5. The specialist charter's explicit model requirement.team-memory-maintenance
When your agent uses it
Merge accepted Squad decisions through the active state backend with exact verification
What it loads into context
Frontmatter
50 tokensAlways available to the agent
SKILL.md
392 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Scope
- Procedure
- Stop Conditions
- Anti-Patterns
Files
SKILL.mdcore2,019 bytes
Read SKILL.md
# Team Memory Maintenance
Use this skill only for accepted durable decisions. The current Scribe role does not maintain
session logs, orchestration logs, specialist histories, onboarding records, or general team memory.
## Scope
- Allowed targets: `.squad/decisions/inbox/` and `.squad/decisions.md`.
- Use the active state backend for mutable Squad state.
- Preserve decision authorship, rationale, and accepted scope.
- Use archival-integrity procedures whenever accepted records are moved or compacted.
## Procedure
work-monitoring
When your agent uses it
Triage Squad work queues without making unintended bulk changes
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
356 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Scope
- Procedure
- Stop Conditions
- Anti-Patterns
Files
SKILL.mdcore1,780 bytes
Read SKILL.md
# Work Monitoring
Use this skill when monitoring Squad issues, pull requests, CI, review feedback, or a labeled work
queue.
## Scope
- Start with a named repository, explicit workstream, label set, and state filter.
- Produce monitoring reports, triage summaries, and narrowly scoped coordination actions.
- Do not turn monitoring into product implementation or broad cleanup.
## Procedure
1. List candidates using the narrowest available filters.{skill-name}
When your agent uses it
{what this skill teaches agents}
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
145 tokensWhen the skill is activated
Reference documents
10,488 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
ceremonies.mddoc1,097 bytescharter.mddoc2,322 bytesconstraint-tracking.mddoc1,192 bytescopilot-instructions.mddoc3,584 bytesissue-lifecycle.mddoc11,524 bytesmcp-config.mddoc2,357 bytesmulti-agent-format.mddoc887 bytesplugin-marketplace.mddoc2,322 bytesralph-reference.mddoc8,828 bytesraw-agent-output.mddoc1,336 bytesroster.mddoc2,676 bytesrouting.mddoc1,808 bytesrun-output.mddoc1,500 bytesscribe-charter.mddoc1,858 bytesskill.mdcore587 bytes
Read SKILL.md
## Context
{When and why this skill applies}
## Patterns
{Specific patterns, conventions, or approaches}
## Examples
{Code examples or references}
## Anti-Patterns
{What to avoid}{skill-name}
When your agent uses it
{what this skill teaches agents}
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
145 tokensWhen the skill is activated
Reference documents
48,671 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
package.jsonralph-triage.jsscripts/notes/fetch.ps1scripts/notes/write-note.ps1
Files
README.mddoc7,063 bytesafter-agent-reference.mddoc5,750 bytescasting-history.jsonasset70 bytescasting-policy.jsonasset823 bytescasting-reference.mddoc7,679 bytescasting-registry.jsonasset146 bytescasting/Futurama.jsonasset108 bytesceremonies.mddoc1,882 bytesceremony-reference.mddoc3,162 bytescharter.mddoc2,053 bytesclient-compatibility-reference.mddoc3,758 bytesconstraint-tracking.mddoc1,192 bytescooperative-rate-limiting.mddoc7,222 bytescopilot-agent.mddoc3,777 bytescopilot-instructions.mddoc3,667 bytesfact-checker-charter.mddoc2,621 bytesfact-checker-policy.mddoc6,104 byteshistory.mddoc255 bytesidentity/now.mddoc209 bytesidentity/wisdom.mddoc384 bytesissue-lifecycle.mddoc11,524 byteskeda-scaler.mddoc5,732 bytesloop.mddoc1,378 bytesmachine-capabilities.mddoc2,415 bytesmcp-config.mddoc3,129 bytesmodel-selection-reference.mddoc16,277 bytesmulti-agent-format.mddoc887 bytesnotes-protocol.mddoc5,513 bytesorchestration-log.mddoc1,104 bytespackage.jsonruns28 bytesplugin-marketplace.mddoc2,437 bytesprd-intake.mddoc3,311 bytesrai-charter.mddoc2,941 bytesrai-policy.mddoc4,057 bytesralph-charter.mddoc2,236 bytesralph-circuit-breaker.mddoc10,439 bytesralph-instructions.mddoc3,445 bytesralph-reference.mddoc8,828 bytesralph-triage.jsruns16,280 bytesraw-agent-output.mddoc1,336 bytesroster.mddoc2,389 bytesrouting.mddoc1,789 bytesrun-output.mddoc1,500 bytesscribe-charter.mddoc3,001 bytesscripts/notes/fetch.ps1runs3,977 bytesscripts/notes/write-note.ps1runs5,449 bytessession-init-reference.mddoc7,194 bytesskill.mdcore587 bytesspawn-reference.mddoc9,351 bytessquad.agent.md.templateasset80,210 bytesworkflow-wiring-appendix-a-code-reviewer.mddoc6,192 bytesworkflow-wiring-appendix-b-documenter.mddoc6,209 bytesworkflow-wiring-guide.mddoc14,233 bytesworkflows/squad-ci.ymlasset506 bytesworkflows/squad-docs.ymlasset1,264 bytesworkflows/squad-heartbeat.ymlasset4,857 bytesworkflows/squad-insider-release.ymlasset2,307 bytesworkflows/squad-issue-assign.ymlasset9,838 bytesworkflows/squad-label-enforce.ymlasset6,935 bytesworkflows/squad-preview.ymlasset1,842 bytesworkflows/squad-promote.ymlasset4,363 bytesworkflows/squad-release.ymlasset2,722 bytesworkflows/squad-triage.ymlasset14,752 bytesworkflows/sync-squad-labels.ymlasset9,408 bytesworktree-reference.mddoc7,553 bytes
Read SKILL.md
## Context
{When and why this skill applies}
## Patterns
{Specific patterns, conventions, or approaches}
## Examples
{Code examples or references}
## Anti-Patterns
{What to avoid}agent-collaboration
When your agent uses it
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
440 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore2,010 bytes
Read SKILL.md
## Context
Every agent on the team follows identical collaboration patterns for worktree awareness, decision recording, and cross-agent communication. These were previously duplicated in every charter's Collaboration section (~300 bytes × 18 agents = ~5.4KB of redundant context). Now centralized here.
The coordinator's spawn prompt already instructs agents to read decisions.md and their history.md. This skill adds the patterns for WRITING decisions and requesting help.
## Patterns
### Worktree Awareness
Use the `TEAM ROOT` path provided in your spawn prompt. All `.squad/` paths are relative to this root. If TEAM ROOT is not provided (rare), run `git rev-parse --show-toplevel` as fallback. Never assume CWD is the repo root.
### Decision Recording
After making a decision that affects other team members, write it to:
`.squad/decisions/inbox/{your-name}-{brief-slug}.md`agent-conduct
When your agent uses it
Shared hard rules enforced across all squad agents
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
283 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore1,345 bytes
Read SKILL.md
## Context
Every squad agent must follow these two hard rules. They were previously duplicated in every charter. Now they live here as a shared skill, loaded once.
## Patterns
### Product Isolation Rule (hard rule)
Tests, CI workflows, and product code must NEVER depend on specific agent names from any particular squad. "Our squad" must not impact "the squad." No hardcoded references to cast-specific agent names in test assertions, CI configs, or product logic. Use generic/parameterized values. If a test needs agent names, use obviously-fake test fixtures (e.g., "test-agent-1", "TestBot").
### Peer Quality Check (hard rule)
Before finishing work, verify your changes don't break existing tests. Run the test suite for files you touched. If CI has been failing, check your changes aren't contributing to the problem. When you learn from mistakes, update your history.md.
## Anti-Patterns
- Don't hardcode dev team agent names in product code or testsarchitectural-proposals
When your agent uses it
How to write comprehensive architectural proposals that drive alignment before code is written
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
1,427 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,335 bytes
Read SKILL.md
## Context
Proposals create alignment before code is written. Cheaper to change a doc than refactor code. Use this pattern when:
- Architecture shifts invalidate existing assumptions
- Product direction changes require new foundation
- Multiple waves/milestones will be affected by a decision
- External dependencies (Copilot CLI, SDK APIs) change
## Patterns
### Proposal Structure (docs/proposals/)
**Required sections:**
1. **Problem Statement** — Why current state is broken (specific, measurable evidence)ci-validation-gates
When your agent uses it
Defensive CI/CD patterns: semver validation, token checks, retry logic, and draft detection
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,072 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Known Failure Modes
- Anti-Patterns
Scripts executed separately
../../../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,562 bytes../../../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
CI workflows must be defensive. These patterns capture lessons from prior release incidents;
they are maintained as version-agnostic gates rather than claims about a particular repository
release.
## Patterns
### Semver Validation Gate
Every publish workflow MUST validate version format before `npm publish`. 4-part versions (e.g., 0.8.21.4) are NOT valid semver — npm mangles them.
```yaml
- name: Validate semver
run: |When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
491 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Checklist — Adding a New CLI Command
- Wiring Patterns by Command Type
- Common Import Pattern
- History
Files
SKILL.mdcore1,844 bytes
Read SKILL.md
# Skill: CLI Command Wiring
**Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.
## Checklist — Adding a New CLI Command
1. **Create command file** in `packages/squad-cli/src/cli/commands/<name>.ts`
- Export a `run<Name>(cwd, options)` async function (or class with static methods for utility modules)
2. **Add routing block** in `packages/squad-cli/src/cli-entry.ts` inside `main()`:
```ts
if (cmd === '<name>') {
const { run<Name> } = await import('./cli/commands/<name>.js');
// parse args, call function
await run<Name>(process.cwd(), options);client-compatibility
When your agent uses it
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,289 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,719 bytes
Read SKILL.md
## Context
Squad runs on multiple Copilot surfaces (CLI, VS Code, JetBrains, GitHub.com). The coordinator must detect its platform and adapt spawning behavior accordingly. Different tools are available on different platforms, requiring conditional logic for agent spawning, SQL usage, and response timing.
## Patterns
### Platform Detection
Before spawning agents, determine the platform by checking available tools:
1. **CLI mode** — `task` tool is available → full spawning control. Use `task` with `agent_type`, `mode`, `model`, `description`, `prompt` parameters. Collect results via `read_agent`.
2. **VS Code mode** — `runSubagent` or `agent` tool is available → conditional behavior. Use `runSubagent` with the task prompt. Drop `agent_type`, `mode`, and `model` parameters. Multiple subagents in one turn run concurrently (equivalent to background mode). Results return automatically — no `read_agent` needed.
coordinator-init-mode
When your agent uses it
The complete two-phase Init Mode protocol the Squad coordinator runs when no team exists yet in the current repo. Phase 1 = propose the team (no files created, wait for user confirm). Phase 2 = create .squad/ scaffolding, casting state, .gitattributes for merge drivers, and the always-on built-ins (Scribe, Ralph, Rai, Fact Checker). Loaded on demand when the coordinator detects no .squad/team.md exists.
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
1,937 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Phase 1: Propose the Team
- Phase 2: Create the Team
Files
SKILL.mdcore7,868 bytes
Read SKILL.md
> **Load this skill when:** you detect that no `.squad/team.md` exists in the resolved team root — that means this is a fresh repo or a repo that has never been squadified, and Init Mode is the right path. The short stub in `squad.agent.md` tells you to load this skill; the full two-phase protocol lives here.
>
> **⚠<U+FE0F> Eager-execution exception:** Init Mode is the ONE exception to the eager-execution / parallel-fan-out doctrine. Phase 1 MUST end with a user confirmation before any file is created. Do not bypass.
## Phase 1: Propose the Team
No team exists yet. **Propose one — but DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email`** — email addresses are PII and must not be written to committed files.
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see the canonical Casting reference at `.squad/templates/casting-reference.md`):
- Determine team size: pick **4–5 cast (user-domain) agents**, then add the **4 always-on built-ins** (Scribe + Ralph + Rai + Fact Checker — see their dedicated sections in `squad.agent.md`). A typical fresh squad has **8–9 total roster entries**, not 4–5.
- Determine assignment shape from the user's project description.
- Derive resonance signals from the session and repo context.coordinator-response-mode
When your agent uses it
Selecting WHO handles work is the Routing table; selecting HOW they handle it (Direct, Lightweight, Standard, Full) is Response Mode. This skill contains the complete decision table, exemplar prompts for each mode, the Lightweight spawn template, and the upgrade rules. Squad coordinator loads this on demand once routing has identified the agent — to pick the right ceremony level for the task.
What it loads into context
Frontmatter
139 tokensAlways available to the agent
SKILL.md
1,172 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Response Mode Selection
- Direct Mode exemplars
- Lightweight Mode exemplars
- Standard Mode exemplars
- Full Mode exemplars
- Mode upgrade rules
- Lightweight Spawn Template
Files
SKILL.mdcore5,018 bytes
Read SKILL.md
> **Load this skill when:** you have routed work to an agent and need to pick the response mode (Direct / Lightweight / Standard / Full). The 1-line stub in `squad.agent.md` is for awareness; this skill is the full decision table + templates.
## Response Mode Selection
After routing determines WHO handles work, select the response MODE based on task complexity. **Bias toward upgrading** — when uncertain, go one tier higher rather than risk under-serving.
| Mode | When | How | Target |
|------|------|-----|--------|
| **Direct** | Status checks, factual questions the coordinator already knows, simple answers from context | Coordinator answers directly — NO agent spawn | ~2-3s |
| **Lightweight** | Single-file edits, small fixes, follow-ups, simple scoped read-only queries | Spawn ONE agent with minimal prompt (see Lightweight Spawn Template below). Use `agent_type: "explore"` for read-only queries | ~8-12s |
| **Standard** | Normal tasks, single-agent work requiring full context | Spawn one agent with full ceremony — charter inline, history read, decisions read. This is the current default | ~25-35s |
| **Full** | Multi-agent work, complex tasks touching 3+ concerns, "Team" requests | Parallel fan-out, full ceremony, Scribe included | ~40-60s |
## Direct Mode exemplarscoordinator-source-of-truth
When your agent uses it
The complete file-by-file source-of-truth hierarchy for Squad: which files are authoritative, which are derived/append-only, who may write each one, who may read each one, and the precedence rules when they conflict. Squad coordinator loads this on demand when it needs to resolve a write conflict, decide where a piece of state belongs, or answer a 'who owns this file' question.
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
1,380 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- State backend note
- File hierarchy
- Rules
Files
SKILL.mdcore5,941 bytesSKILL.mdcore5,941 bytes
Read SKILL.md
> **Load this skill when:** the coordinator (or any agent) needs to resolve a "where does this state belong?" or "who is allowed to write this file?" question — e.g., when about to write `.squad/decisions.md`, when reviewing whether an agent broke the append-only rule, when answering a user question about Squad's file layout, or when adjudicating a conflict between two files. The short summary in `squad.agent.md` is for routing; this skill is the full reference.
## State backend note
Files below marked as **"Derived / append-only"** are **mutable state** — agents access them with runtime state tools (`squad_state_read`, `squad_state_write`, `squad_state_append`, `squad_state_create_if_absent`, `squad_state_delete`, `squad_state_list`). The runtime decides whether the configured backend stores them on disk, git-native state, or an external provider. Files marked as **"Authoritative"** are **static config** and always live on disk regardless of backend.
## File hierarchy
| File | Status | Who May Write | Who May Read |
|------|--------|---------------|--------------|
| `.github/agents/squad.agent.md` | **Authoritative governance.** All roles, handoffs, gates, and enforcement rules. | Repo maintainer (human) | Squad (Coordinator) |
| `.squad/decisions.md` | **Authoritative decision ledger.** Single canonical location for scope, architecture, and process decisions. | Squad (Coordinator) — append only | All agents |
| `.squad/team.md` | **Authoritative roster.** Current team composition. | Squad (Coordinator) | All agents |
| `.squad/routing.md` | **Authoritative routing.** Work assignment rules. | Squad (Coordinator) | Squad (Coordinator) |cross-machine-coordination
When your agent uses it
Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002).
What it loads into context
Frontmatter
58 tokensAlways available to the agent
SKILL.md
3,211 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Usage
- File Formats
- Security Model
- Configuration
- Examples
- Error Handling
- Monitoring & Debugging
- Integration with Ralph Watch
- Migration from Manual Handoff
- Future Enhancements
- Questions?
Files
SKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytes
Read SKILL.md
# Skill: Cross-Machine Coordination Pattern
**Skill ID:** `cross-machine-coordination`
**Owner:** Ralph (Work Monitor)
**Squad Integration:** All agents
**Status:** Design proposal — NOT implemented
> ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
> the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
> and `squad watch` do not poll task files, run whitelisted commands, or
> write result files. Everything below describes a proposed design, not
> current behavior. Do not tell users it works, and do not create task files
> expecting another machine to pick them up.
>cross-squad
When your agent uses it
Coordinating work across multiple Squad instances — discovery, delegation, and disambiguation when the user says 'squad' (the product) vs casual English 'group of agents'.
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
2,402 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Disambiguation: "squad" vs ad-hoc agents
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytes
Read SKILL.md
## Context
> **Read this FIRST any time the user says "squad" as a thing to spawn, delegate to, address, or fan out to** — e.g., *"spawn two squads of designers and devs"*, *"ask the other squad"*, *"delegate to a squad"*. In Squad-PRODUCT vocabulary, "squad" is a **peer** (an independent installation with its own `.squad/`, `team.md`, MCP server, and agents) — NOT a generic English synonym for "team" or "group". Do not fan out raw `task` agents inside your own coordinator context when the user means "another squad". Use the discovery and communication patterns below (and the companion `cross-squad-communication` skill for the actual protocols).
When an organization runs multiple Squad instances (e.g., platform-squad, frontend-squad, data-squad), those squads need to discover each other, share context, and hand off work across repository boundaries. This skill teaches agents how to coordinate across squads without creating tight coupling.
> **Companion skill — for protocol details:** `cross-squad-communication/SKILL.md` covers the four communication patterns (synchronous CLI, read-only knowledge query, git-based async, and GitHub-issue-based delegation) once a peer squad is discovered via the registry below. This skill answers "who?" — the companion answers "how?".
Cross-squad orchestration applies when:
- A task requires capabilities owned by another squad
- An architectural decision affects multiple squads
- A feature spans multiple repositories with different squads
- A squad needs to request infrastructure, tooling, or support from another squad
cross-squad-communication
When your agent uses it
Protocol for sending queries, delegating tasks, and sharing context between independent Squad instances across different repositories
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
4,739 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Universal rule: every `copilot` spawn into a peer squad MUST pass `--agent squad`
- Discovery Protocol
- Platform Compatibility Matrix
- Examples
- Anti-Patterns
- Validation Status
Files
SKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytes
Read SKILL.md
## Context
When multiple repositories each have their own Squad (AI team), they need to exchange information: knowledge queries, PR reviews, task delegation, and dependency analysis. Each squad has its own agents, MCP tools, and issue tracker — there is no shared runtime.
> **Companion skill — read first:** `cross-squad/SKILL.md` covers **discovery** of peer squads via `squad registry add/list/remove`. This skill picks up after a peer is known and covers the **communication protocols** themselves — the four numbered patterns below: Pattern 0 (synchronous CLI), Pattern 1 (read-only knowledge query), Pattern 2 (git-based async), and Pattern 3 (GitHub-issue-based delegation). A separate non-numbered appendix (Cross-Repo Dependency Scan) is provided as a related analysis tool, not a communication pattern. The two skills are designed to be used together.
**When this skill applies:**
- A squad agent needs information from another squad-enabled repo
- A task needs to be delegated to another squad
- Cross-repo dependency analysis is needed
- PR review requests span repo boundaries
**Key constraint:** Each squad has its own runtime, MCP tools, and issue tracker. Cross-squad communication can be **synchronous** (via CLI session targeting the other repo) or **asynchronous** (file-based or issue-based). The coordinator decides which approach fits.
distributed-mesh
When your agent uses it
How to coordinate with squads on different machines using git as transport
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
3,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Patterns
- Examples
- AGENT WORKFLOW (Deterministic Setup)
- Anti-Patterns
Scripts executed separately
sync-mesh.ps1sync-mesh.sh
Files
SKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesmesh.json.exampleasset762 bytessync-mesh.ps1runs3,861 bytessync-mesh.shruns3,395 bytes
Read SKILL.md
## SCOPE
**✅ THIS SKILL PRODUCES (exactly these, nothing more):**
1. **`mesh.json`** — Generated from user answers about zones and squads (which squads participate, what zone each is in, paths/URLs for each), using `mesh.json.example` in this skill's directory as the schema template
2. **`sync-mesh.sh` and `sync-mesh.ps1`** — Copied from this skill's directory into the project root (these are bundled resources, NOT generated code)
3. **Zone 2 state repo initialization** (if applicable) — If the user specified a Zone 2 shared state repo, run `sync-mesh.sh --init` to scaffold the state repo structure
4. **A decision entry** in `.squad/decisions/inbox/` documenting the mesh configuration for team awareness
**❌ THIS SKILL DOES NOT PRODUCE:**
- **No application code** — No validators, libraries, or modules of any kind
- **No test files** — No test suites, test cases, or test scaffolding
- **No GENERATING sync scripts** — They are bundled with this skill as pre-built resources. COPY them, don't generate them.docs-standards
When your agent uses it
Microsoft Style Guide + Squad-specific documentation patterns
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
524 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore2,270 bytesSKILL.mdcore2,270 bytes
Read SKILL.md
## Context
Squad documentation follows the Microsoft Style Guide with Squad-specific conventions. Consistency across docs builds trust and improves discoverability.
## Patterns
### Microsoft Style Guide Rules
- **Sentence-case headings:** "Getting started" not "Getting Started"
- **Active voice:** "Run the command" not "The command should be run"
- **Second person:** "You can configure..." not "Users can configure..."
- **Present tense:** "The system routes..." not "The system will route..."
- **No ampersands in prose:** "and" not "&" (except in code, brand names, or UI elements)
### Squad Formatting Patternse2e-template-testing
When your agent uses it
End-to-end validation of coordinator and agent template changes
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
6,164 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- When To Use
- Prerequisites
- Workflow
- Progress Reporting
- Test Matrix Template
- Tips
- Fast-Fail Rules
- PII Protection — Mandatory
- Anti-Patterns
- Sandbox / Permission Notes
- Confidence
Files
SKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytes
Read SKILL.md
## Context
Squad's coordinator prompt (`squad.agent.md`) and agent charters (e.g.
`scribe-charter.md`) are shipped as templates in `.squad-templates/`. Changes to
these files affect how every squad session behaves — but unit tests can't catch
prompt-level regressions because the prompts are interpreted by an LLM at
runtime.
This skill describes how to validate template changes end-to-end by running real
squad sessions against a locally-built CLI that includes your modified templates.
## When To Use
- You changed `.squad-templates/squad.agent.md` (coordinator prompt)economy-mode
When your agent uses it
Shifts Layer 3 model selection to cost-optimized alternatives when economy mode is active.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,396 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Activation Methods
- Economy Model Selection Table
- AGENT WORKFLOW
- Config Schema
- Anti-Patterns
Files
SKILL.mdcore5,547 bytes
Read SKILL.md
## SCOPE
✅ THIS SKILL PRODUCES:
- A modified Layer 3 model selection table applied when economy mode is active
- `economyMode: true` written to `.squad/config.json` when activated persistently
- Spawn acknowledgments with `💰` indicator when economy mode is active
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Cost reports or billing artifacts
- Changes to Layer 0, Layer 1, or Layer 2 resolution (user intent always wins)
## Context
error-recovery
When your agent uses it
Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
785 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Retry with Backoff
- 2. Fallback Alternatives
- 3. Diagnose-and-Fix
- 4. Escalate with Context
- 5. Graceful Degradation
- Applying These Patterns
Files
SKILL.mdcore3,466 bytesSKILL.mdcore3,466 bytes
Read SKILL.md
# Error Recovery Patterns
Standard recovery patterns for all squad agents. When something fails, **adapt** — don't just report the failure.
---
## 1. Retry with Backoff
**When:** Transient failures — API timeouts, rate limits, network errors, temporary service unavailability.
**Pattern:**
1. Wait briefly, then retry (start at 2s, double each attempt)
2. Maximum 3 retries before escalating
3. Log each attempt with the error receivedexternal-comms
When your agent uses it
Workflow for scanning, drafting, and presenting community responses with a human review gate
What it loads into context
Frontmatter
228 tokensAlways available to the agent
SKILL.md
3,012 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore13,159 bytesSKILL.mdcore13,159 bytes
Read SKILL.md
## Context
Phase 1 is **draft-only mode**.
- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.
- **Human review gate is mandatory** — the communications owner never posts autonomously.
- Every action is logged to `.squad/comms/audit/`.
- This workflow is triggered manually only ("check community") — no automated activation in Phase 1.
## Patterns
### 1. Scan
Find unanswered community items with GitHub MCP tools first, or `gh issue list` / `gh api` as fallback for issues and discussions.fact-checking
When your agent uses it
Review and validate claims using counter-hypothesis testing. Use when verifying technical content, checking references, validating API endpoints, or performing quality assurance on deliverables.
What it loads into context
Frontmatter
97 tokensAlways available to the agent
SKILL.md
469 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore2,046 bytes
Read SKILL.md
# Skill: Fact Checking
## Context
Codifies the challenger agent review output format and methodology so any agent performing fact-checking or review produces consistent, structured output.
## Pattern
### Review Methodology
For every claim or deliverable under review:
1. Ask: "What evidence supports this? What would disprove it?"
2. Generate counter-hypotheses and test them against available data
3. Verify URLs, package names, API endpoints, and external references actually exist
4. Flag confidence levels: ✅ Verified, ⚠<U+FE0F> Unverified, ❌ Contradictedgh-auth-isolation
When your agent uses it
Safely manage multiple GitHub identities (EMU + personal) in agent workflows
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
1,543 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytes
Read SKILL.md
## Context
Many developers use GitHub through an Enterprise Managed User (EMU) account at work while maintaining a personal GitHub account for open-source contributions. AI agents spawned by Squad inherit the shell's default `gh` authentication — which is usually the EMU account. This causes failures when agents try to push to personal repos, create PRs on forks, or interact with resources outside the enterprise org.
This skill teaches agents how to detect the active identity, switch contexts safely, and avoid mixing credentials across operations.
## Patterns
### Detect Current Identity
Before any GitHub operation, check which account is active:
```bash
gh auth statusgh-aw-enlistment
When your agent uses it
Enlist a repository into Squad by installing the supported GitHub Agentic Workflows (gh-aw) bootstrap — with strict compilation, an explicit safe-update allowlist, and a human-reviewed bootstrap PR. Never auto-merges.
What it loads into context
Frontmatter
308 tokensAlways available to the agent
SKILL.md
7,239 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytes
Read SKILL.md
## Context
Use this skill when a user wants to **enlist a repository into Squad** through
**GitHub Agentic Workflows (`gh aw`)** — e.g. "set up Squad agentic workflows",
"enlist this repo in Squad", or "install Squad gh-aw workflows". The end state is
the `/squad` slash command being live on the repo, delivered as a **human-reviewed
bootstrap pull request** — never an auto-merge.
This skill **operationalizes** the supported bootstrap path. It does not summarize
it: each step below is a gate with explicit success evidence and **STOP conditions**.
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this
skill ever disagree, the guide wins — re-read it before proceeding.
git-workflow
When your agent uses it
Squad branching model: dev-first workflow with insiders preview channel
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,581 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Branch Naming Convention
- Workflow for Issue Work
- Parallel Multi-Issue Work (Worktrees)
- Multi-Repo Downstream Scenarios
- Anti-Patterns
- Promotion Pipeline
Files
SKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytes
Read SKILL.md
## Context
Squad uses a three-branch model. **All feature work starts from `dev`, not `main`.**
| Branch | Purpose | Publishes |
|--------|---------|-----------|
| `main` | Released, tagged, in-npm code only | `npm publish` on tag |
| `dev` | Integration branch — all feature work lands here | `npm publish --tag preview` on merge |
| `insiders` | Early-access channel — synced from dev | `npm publish --tag insiders` on sync |
## Branch Naming Convention
Issue branches MUST use: `squad/{issue-number}-{kebab-case-slug}`
github-multi-account
When your agent uses it
Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
What it loads into context
Frontmatter
80 tokensAlways available to the agent
SKILL.md
1,038 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Activate
- What to Do (as the AI agent)
- After Setup — Usage Rules
- Repo-Specific Account Binding
- For Squad Agents
Files
SKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytes
Read SKILL.md
# GitHub Multi-Account — AI-Driven Setup
## When to Activate
When the user has multiple GitHub accounts (check with `gh auth status`). If you see 2+ accounts listed, this skill applies.
## What to Do (as the AI agent)
### Step 1: Detect accounts
Run: `gh auth status`
Look for multiple accounts. Note which usernames are listed.
### Step 2: Ask the user
Ask: "I see you have multiple GitHub accounts: {list them}. Which one is your personal account and which is your work/EMU account?"
history-hygiene
When your agent uses it
Record reconciled final outcomes to history.md, not intermediate requests or obsolete claims
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
545 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Patterns
- Examples
- Stop conditions
- Anti-Patterns
Files
SKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytes
Read SKILL.md
## Context
History files are cold-start context, not the source of truth for live behavior. Stale or incorrect
entries poison decision-making downstream. A stale-history incident proved that an old history claim
can cause future spawns to repeat a reversed decision. Reconcile an operational claim with current
implementation, workflows, tests, documentation, skills, and applicable decisions before acting on it.
## Scope
- Activate when recording, correcting, or promoting finalized operational history.
- Allowed targets: the explicitly selected history file and a focused reusable skill when the rule
applies beyond one role.
- Exclusions: intermediate requests, stale assignments, roster metadata, and unrelated cleanup.
humanizer
When your agent uses it
Tone enforcement patterns for external-facing community responses
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,014 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,430 bytes
Read SKILL.md
## Context
Use this skill whenever the communications owner drafts external-facing responses for issues or discussions.
- Tone must be warm, helpful, and human-sounding — never robotic or corporate.
- The repository's communication policy applies everywhere: **Humanized tone is mandatory**.
- This applies to **all external-facing content** drafted by the communications owner in Phase 1 issues/discussions workflows.
## Patterns
1. **Warm opening** — Start with acknowledgment ("Thanks for reporting this", "Great question!")
2. **Active voice** — "We're looking into this" not "This is being investigated"
3. **Second person** — Address the person directly ("you" not "the user")
4. **Conversational connectors** — "That said...", "Here's what we found...", "Quick note:"init-mode
When your agent uses it
Team initialization flow (Phase 1 proposal + Phase 2 creation)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
1,749 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,100 bytesSKILL.mdcore7,100 bytes
Read SKILL.md
## Context
Init Mode activates when `.squad/team.md` does not exist, or exists but has zero roster entries under `## Members`. The coordinator proposes a team (Phase 1), waits for user confirmation, then creates the team structure (Phase 2).
## Patterns
### Phase 1: Propose the Team
No team exists yet. Propose one — but **DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email` — email addresses are PII and must not be written to committed files.**
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see that section):
- Determine team size (typically 4–5 + Scribe).iterative-retrieval
When your agent uses it
Max-3-cycle protocol for agent sub-tasks with WHY context and coordinator validation. Use when spawning sub-agents to complete scoped work.
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,385 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Spawn Prompt Template
- 3-Cycle Protocol
- Coordinator Validation Checklist
- When to Escalate vs Retry
- Issue Dedup Check (Mandatory)
- Mandatory Output Requirement (Research-Then-Execute)
- Anti-Patterns
- Examples
Files
SKILL.mdcore6,341 bytesSKILL.mdcore6,341 bytes
Read SKILL.md
# Iterative Retrieval Skill
Squad agents frequently spawn sub-agents to complete scoped work. Without structure, these
handoffs become vague, cycles multiply, and outputs land without being checked. The
**Iterative Retrieval Pattern** caps cycles at 3, mandates WHY context in every spawn, and
requires the coordinator to validate agent output before closing an issue.
---
## Spawn Prompt Template
Every agent spawn must include the following four sections. Copy and fill in the template:
```model-selection
When your agent uses it
Determines which LLM model to use for each agent spawn
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- 5-Layer Model Resolution Hierarchy
- AGENT WORKFLOW
- Config Schema
- Fallback Chains
Files
SKILL.mdcore8,082 bytes
Read SKILL.md
# Model Selection
> Determines which LLM model to use for each agent spawn.
## SCOPE
✅ THIS SKILL PRODUCES:
- A resolved `model` parameter for every `task` tool call
- Persistent model preferences in `.squad/config.json`
- Spawn acknowledgments that include the resolved model
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Model performance benchmarksnap
When your agent uses it
Context hygiene — compress, prune, archive .squad/ state
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
208 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What It Does
- When To Use
- Invocation
- Confidence
Files
SKILL.mdcore792 bytes
Read SKILL.md
# Skill: nap
> Context hygiene — compress, prune, archive .squad/ state
## What It Does
Reclaims context window budget by compressing agent histories, pruning old logs,
archiving stale decisions, and cleaning orphaned inbox files.
## When To Use
- Before heavy fan-out work (many agents will spawn)
- When history.md files exceed 15KB
- When .squad/ total size exceeds 1MBnotification-routing
When your agent uses it
Route agent notifications to specific channels by type — prevent alert fatigue from single-channel flooding
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
730 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
- Distributed Systems Pattern
Files
SKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytes
Read SKILL.md
## Context
When a Squad grows beyond a few agents, notifications flood a single channel — failure alerts drown in daily
briefings, tech news buries security findings, and everything gets ignored. This is the pub-sub problem:
a single message queue for everything is a recipe for missed alerts.
The fix is **topic-based routing**: agents tag notifications with a channel type, and a routing function
sends them to the appropriate destination.
**Trigger symptoms:**
- Important alerts missed because they're buried in routine notifications
- Team members turning off notifications entirely (signal overwhelm)
- Onboarding friction: "where do I look for X?"
pr-review-response
When your agent uses it
Teaches agents to reply to PR review comment threads after fixing issues, making resolutions traceable
What it loads into context
Frontmatter
154 tokensAlways available to the agent
SKILL.md
2,574 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- SCOPE
- Patterns
- AGENT WORKFLOW (Summary)
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytes
Read SKILL.md
## Context
When an agent fixes code in response to PR review comments (from Copilot, a human reviewer, or any GitHub reviewer), the fix alone is not enough. The reviewer needs to see — on the PR thread itself — which comments were addressed and how. Without replies, comments stay visually unresolved, reviewers must re-read the entire diff to verify fixes, and there's no traceable link between feedback and resolution.
Use this skill whenever:
- You are fixing code based on PR review feedback
- You are addressing Copilot review suggestions
- You are responding to reviewer-requested changes on a PR
- A squad member hands you review comments to resolve
## SCOPE
✅ THIS SKILL PRODUCES:
- Reply comments on each review thread explaining the fixpr-screenshots
When your agent uses it
Capture Playwright screenshots and embed them in GitHub PR descriptions
What it loads into context
Frontmatter
62 tokensAlways available to the agent
SKILL.md
1,065 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../../../scripts/capture-pr-screenshots.mjs
Files
SKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytes../../../../../scripts/capture-pr-screenshots.mjsruns1,855 bytes
Read SKILL.md
## Context
When a PR includes visual changes (docs sites, UI components, generated pages), reviewers
need to see what the PR delivers without checking out the branch. Screenshots belong in
the **PR description body**, not as committed files and not as text descriptions.
Use this skill whenever:
- A PR touches docs site pages (Astro, Starlight, etc.)
- A PR adds or changes UI components
- A PR generates visual artifacts (TypeDoc, Storybook, diagrams)
- Playwright tests already capture screenshots as part of testing
## Patterns
project-conventions
When your agent uses it
Core conventions and patterns for this codebase
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,555 bytes
Read SKILL.md
## Context
> **This is a starter template.** Replace the placeholder patterns below with your actual project conventions. Skills train agents on codebase-specific practices — accurate documentation here improves agent output quality.
## Patterns
### [Pattern Name]
Describe a key convention or practice used in this codebase. Be specific about what to do and why.
### Error Handling
<!-- Example: How does your project handle errors? -->
<!-- - Use try/catch with specific error types? -->ralph-two-pass-scan
When your agent uses it
Cuts GitHub API calls by separating lightweight list scanning from full hydration
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
345 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore1,336 bytes
Read SKILL.md
# Skill: Ralph — Two-Pass Issue Scanning
**Confidence:** high
**Domain:** work-monitoring
**Last validated:** 2026-03-24
## Context
Cuts GitHub API calls from N+1 to ~7 per round (~72% reduction) by separating list scanning from full hydration.
Addresses the scanning inefficiency described in issue #596.
## Pattern
### Pass 1 — Lightweight Scan
```reflect
When your agent uses it
Learning capture system that extracts HIGH/MED/LOW confidence patterns from conversations to prevent repeating mistakes. Use after user corrections ("no", "wrong"), praise ("perfect", "exactly"), or when discovering edge cases. Complements .squad/agents/{agent}/history.md and .squad/decisions.md.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Integration with Squad Architecture
- Triggers
- Process
- Usage Examples
- When to Use
- See Also
Files
SKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytes
Read SKILL.md
# Reflect Skill
**Critical learning capture system** for Squad. Prevents repeating mistakes and preserves successful patterns across sessions.
Analyze conversations and propose improvements to squad knowledge based on what worked, what didn't, and edge cases discovered. **Every correction is a learning opportunity.**
---
## Integration with Squad Architecture
**Reflect complements existing Squad knowledge systems:**
1. **Governed runtime memory** — Durable observations and decisions persisted through configured state tools
2. **Tracked governing source** — Coordinator prompts, skills, and other behavior-defining files changed through focused pull requestsrelease-process
When your agent uses it
Prepare, publish, recover, and verify Squad insider, preview, and stable releases
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
2,578 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Non-negotiable rules
- Required credentials
- Human release-trigger boundary
- Prepare a release
- Publish a preview release
- Publish an insider release
- Publish a stable release
- Verify publication
- Recovery
- After a stable release
Files
SKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytes
Read SKILL.md
# Release Process
This is the canonical Squad release runbook. Squad has three release channels
and two long-lived branches:
| Source | Version | GitHub | npm | Standalone | Homebrew/WinGet |
|--------|---------|--------|-----|------------|-----------------|
| Insider dispatch from `dev` | Generated `X.Y.Z-insider.N` | Prerelease | `insider` | Yes | Yes |
| Release dispatch from `dev` | `X.Y.Z-preview.N` | Prerelease | `preview` | Yes | Yes |
| Promotion PR merged to `main` (or legacy workflow push) | `X.Y.Z` | Stable/latest | `latest` | Yes | Yes |
The repository does not use a staging `preview` branch. Preview is a release
channel, not a branch.
reskill
When your agent uses it
Extract durable operating knowledge before slimming or replacing team charters
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,223 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Process
- Patterns
- Anti-Patterns
Files
SKILL.mdcore5,760 bytesSKILL.mdcore5,760 bytes
Read SKILL.md
## Context
When the coordinator hears "team, reskill" (or similar: "optimize context", "slim down charters"), trigger a team-wide optimization pass. The goal: reduce per-agent context consumption by extracting shared patterns from charters and histories into reusable skills.
This is a periodic maintenance activity. Run whenever charter/history bloat is suspected or a
team will be re-cast. Charters and histories are evidence, not the authority for current product
behavior: reconcile candidates against current source, workflows, tests, documentation, skills,
and decisions before preserving them.
## Process
### Step 1: Audit
Read all active agent charters and histories, then inventory existing skills. Measure byte sizes.
Identify:When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
1,352 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Purpose
- Core Function: Test-RetroOverdue
- Coordinator Integration
- Action Item Enforcement
- Cadence Enforcement
- Skill Metadata
Files
SKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytes
Read SKILL.md
# Skill: Retro Enforcement
## Purpose
Ensure retrospectives happen on schedule and that their action items are tracked in GitHub Issues — not markdown checklists.
This skill addresses a specific, measured failure mode: **0% completion rate on markdown retro action items across 6 consecutive retrospectives**. GitHub Issues have an 85%+ completion rate in the same squad. The format was the problem, not the people.
## Core Function: Test-RetroOverdue
```powershell
function Test-RetroOverdue {
param(
[string]$LogDir = ".squad/log",
[int]$WindowDays = 7,reviewer-protocol
When your agent uses it
Reviewer rejection workflow and strict lockout semantics
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
1,009 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,658 bytes
Read SKILL.md
## Context
When a team member has a **Reviewer** role (e.g., Tester, Code Reviewer, Lead), they may approve or reject work from other agents. On rejection, the coordinator enforces strict lockout rules to ensure the original author does NOT self-revise. This prevents defensive feedback loops and ensures independent review.
## Patterns
### Reviewer Rejection Protocol
Quality and test reviewers may block a PR or merge on quality, coverage, or regression grounds. Treat that block as a rejection that enters the protocol below.
When a team member has a **Reviewer** role:
- Reviewers may **approve** or **reject** work from other agents.
- On **rejection**, the Reviewer may choose ONE of:secret-handling
When your agent uses it
Never read .env files or write secrets to .squad/ committed files
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
1,994 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,524 bytesSKILL.mdcore7,524 bytes
Read SKILL.md
## Context
Spawned agents have read access to the entire repository, including `.env` files containing live credentials. If an agent reads secrets and writes them to `.squad/` files (decisions, logs, history), Scribe auto-commits them to git, exposing them in remote history. This skill codifies absolute prohibitions and safe alternatives.
## Patterns
### Prohibited File Reads
**NEVER read these files:**
- `.env` (production secrets)
- `.env.local` (local dev secrets)
- `.env.production` (production environment)
- `.env.development` (development environment)
- `.env.staging` (staging environment)session-recovery
When your agent uses it
Find and resume interrupted Copilot CLI sessions using session_store queries
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,215 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytes
Read SKILL.md
## Context
Squad agents run in Copilot CLI sessions that can be interrupted — terminal crashes, network drops, machine restarts, or accidental window closes. When this happens, in-progress work may be left in a partially-completed state: branches with uncommitted changes, issues marked in-progress with no active agent, or checkpoints that were never finalized.
Copilot CLI stores session history in a SQLite database called `session_store` (read-only, accessed via the `sql` tool with `database: "session_store"`). This skill teaches agents how to query that store to detect interrupted sessions and resume work.
## Patterns
### 1. Find Recent Sessions
Query the `sessions` table filtered by time window. Include the last checkpoint to understand where the session stopped:
```sql
SELECTsquad
When your agent uses it
Squad's command catalog and interactive menu. Invoke via /squad (slash command) or natural language ("squad commands", "what can squad do", "show me squad options"). Presents categorized operations (Install & Upgrade, Team Management, Issues & PRs, Plugins & Skills, Model & Cost, Sessions & State) as an interactive picker. Routes to the right squad CLI command or the Squad coordinator agent.
What it loads into context
Frontmatter
103 tokensAlways available to the agent
SKILL.md
2,680 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,874 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill (via `/squad` slash command, "squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-commands
When your agent uses it
Categorized catalog of common Squad operations. Coordinator reads this file and presents it as an interactive menu when the user asks for available commands or help.
What it loads into context
Frontmatter
82 tokensAlways available to the agent
SKILL.md
2,652 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,748 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill ("squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-conventions
When your agent uses it
Core conventions and patterns used in the Squad codebase
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
824 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore3,440 bytesSKILL.mdcore3,440 bytes
Read SKILL.md
## Context
These conventions apply to all work on the Squad CLI tool (`create-squad`). Squad is a zero-dependency Node.js package that adds AI agent teams to any project. Understanding these patterns is essential before modifying any Squad source code.
## Patterns
### Zero Dependencies
Squad has zero runtime dependencies. Everything uses Node.js built-ins (`fs`, `path`, `os`, `child_process`). Do not add packages to `dependencies` in `package.json`. This is a hard constraint, not a preference.
### Node.js Built-in Test Runner
Tests use `node:test` and `node:assert/strict` — no test frameworks. Run with `npm test`. Test files live in `test/`. The test command is `node --test test/`.
### Error Handling — `fatal()` Pattern
All user-facing errors use the `fatal(msg)` function which prints a red `✗` prefix and exits with code 1. Never throw unhandled exceptions or print raw stack traces. The global `uncaughtException` handler calls `fatal()` as a safety net.
squad-help
When your agent uses it
How to actually use Squad — Squad is a custom Copilot agent (invoked via the task tool with agent_type='Squad'), not a skill. This file explains the right invocation paths for setting up a team, listing squad commands, and initializing Squad in a new project.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,043 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Squad is a custom agent, not a skill
- How to actually use Squad
- What NOT to do
- How this skill was discovered
- See also
Files
SKILL.mdcore4,198 bytesSKILL.mdcore4,198 bytes
Read SKILL.md
# Skill: squad-help
> **Quick reference.** If you're reading this because a user said "use squad" or "squad" or "set up a squad", you're in the right place — read on for the correct invocation paths.
---
## Squad is a custom agent, not a skill
The Squad framework registers a **custom Copilot CLI agent** at `.github/agents/squad.agent.md`. The agent is named **`Squad`** and its description is *"Your AI team. Describe what you're building, get a team of specialists that live in your repo."*
Copilot CLI agents and skills are different things:
| Thing | How to invoke | Example |
|---|---|---|squad-version-check
When your agent uses it
Internals of how @bradygaster/squad-cli stamps its version, how `squad upgrade` works (what it preserves vs overwrites), and how to probe the npm registry for the latest version from a coordinator prompt.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,702 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What This Skill Covers
- Package & Registry Facts
- Version Stamping Mechanism
- `squad upgrade` Behavior
- `.squad/config.json` — What It Holds
- Version-Check Probe (npm Registry)
- Upstream OS-Specific Cache
- Repo-Local Cache Convention: `.squad/.cache/version-check.json`
- Key File Paths (installed CLI)
Files
SKILL.mdcore6,296 bytes
Read SKILL.md
# SKILL: Squad CLI Internals — Version Stamping & Upgrade Mechanics
**Confidence:** medium
**Discovered by:** repository investigation
**Date:** 2026-05-26
**Validated in:** Issue #1173 recon (bradygaster/squad)
---
## What This Skill Covers
Reusable knowledge about how `@bradygaster/squad-cli` stamps its version into `squad.agent.md`, how `squad upgrade` works, what it preserves vs. overwrites, and how to probe the npm registry for the latest version from a coordinator prompt.
---test-discipline
When your agent uses it
Design assertions that prove the actual contract and fail on realistic regressions
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytes
Read SKILL.md
## Context
Use this skill whenever a change affects an API, public interface, workflow contract, generated
artifact, or user-observable behavior. A passing status-only test is not evidence that the
intended artifact or contract is correct.
## Patterns
- Update tests in the same change as an API, signature, export, or public-interface change.
- Make assertions name and inspect the offending artifact, output, or configuration field—not just
an exit code or broad success status.
- Mutation-test critical gates by changing the real source or emitted artifact and proving the
assertion turns red.
- Exercise every caller path for a changed parser, normalizer, or helper; isolated unit coveragetiered-memory
When your agent uses it
Three-tier agent memory model (hot/cold/wiki) for context reduction per spawn
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
1,775 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Memory Tiers
- When to Load Each Tier
- Spawn Template Pattern
- Integration with Scribe Agent (design — not yet implemented)
- Implementation Checklist (tracked in #1264)
- References
- Spawn Template
- Task
- WHY
- Success Criteria
- Memory Context
- Escalation
- Notes
Files
SKILL.mdcore7,506 bytes
Read SKILL.md
# Skill: Tiered Agent Memory
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, and spawn-template tier-aware reads are tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264). Until those land, agents continue to load full `history.md` + `decisions.md` on every spawn.
## Overview
Squad agents today load their full context history on every spawn, which grows unboundedly across sessions. The Tiered Agent Memory model proposes a three-tier separation so agents only load the bytes that are actually relevant to the current task, with older context kept available on demand.
---
## Memory Tiers
### 🔥 Hot Tier — Current Session Context
- **Size target:** keep small (~2–4KB typical)versioning-policy
When your agent uses it
SemVer rules for Squad's stable, preview, insider, and local package versions
What it loads into context
Frontmatter
60 tokensAlways available to the agent
SKILL.md
1,153 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- 1. Supported versions
- 2. Package versions stay in lockstep
- 3. Prerelease workspace dependency rule
- 4. Local build versions are ephemeral
- 5. Release lifecycle
- 6. Ownership
- 7. CI enforcement
- Quick reference
Scripts executed separately
../../../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytes../../../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
Squad publishes `@bradygaster/squad-sdk` and
`@bradygaster/squad-cli` from one npm workspace. Their versions and the root
version move together. A mismatched prerelease dependency can make npm silently
resolve an older registry SDK instead of the local workspace.
## 1. Supported versions
| Use | Version | Committed branch |
|---|---|---|
| Stable release | `MAJOR.MINOR.PATCH` | `main`, and briefly `dev` during promotion |
| Preview release | `MAJOR.MINOR.PATCH-preview.N` | `dev` |
| Insider snapshot | `MAJOR.MINOR.PATCH-insider.N` | Generated by the insider workflow |windows-compatibility
When your agent uses it
Cross-platform path handling and command patterns
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
1,572 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../../../scripts/fix-crlf-worktree.mjs
Files
SKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytes../../../../../scripts/fix-crlf-worktree.mjsruns5,444 bytes
Read SKILL.md
## Context
Squad runs on Windows, macOS, and Linux. Several bugs have been traced to platform-specific assumptions: ISO timestamps with colons (illegal on Windows), `git -C` with Windows paths (unreliable), forward-slash paths in Node.js on Windows.
## Patterns
### Filenames & Timestamps
- **Never use colons in filenames:** ISO 8601 format `2026-03-15T05:30:00Z` is illegal on Windows
- **Use `safeTimestamp()` utility:** Replaces colons with hyphens → `2026-03-15T05-30-00Z`
- **Centralize formatting:** Don't inline `.toISOString().replace(/:/g, '-')` — use the utility
### Git Commands
- **Never use `git -C {path}`:** Unreliable with Windows paths (backslashes, spaces, drive letters)
- **Always `cd` first:** Change directory, then run git commands{skill-name}
When your agent uses it
{what this skill teaches agents}
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
145 tokensWhen the skill is activated
Reference documents
48,360 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
package.jsonralph-triage.jsscripts/notes/fetch.ps1scripts/notes/write-note.ps1
Files
README.mddoc7,063 bytesafter-agent-reference.mddoc5,750 bytescasting-history.jsonasset70 bytescasting-policy.jsonasset823 bytescasting-reference.mddoc7,679 bytescasting-registry.jsonasset146 bytescasting/Futurama.jsonasset108 bytesceremonies.mddoc1,882 bytesceremony-reference.mddoc3,162 bytescharter.mddoc2,053 bytesclient-compatibility-reference.mddoc3,758 bytesconstraint-tracking.mddoc1,192 bytescooperative-rate-limiting.mddoc7,222 bytescopilot-agent.mddoc3,777 bytescopilot-instructions.mddoc3,667 bytesfact-checker-charter.mddoc2,621 bytesfact-checker-policy.mddoc6,104 byteshistory.mddoc255 bytesidentity/now.mddoc209 bytesidentity/wisdom.mddoc384 bytesissue-lifecycle.mddoc11,524 byteskeda-scaler.mddoc5,732 bytesmachine-capabilities.mddoc2,415 bytesmcp-config.mddoc3,129 bytesmodel-selection-reference.mddoc16,277 bytesmulti-agent-format.mddoc887 bytesnotes-protocol.mddoc5,513 bytesorchestration-log.mddoc1,104 bytespackage.jsonruns28 bytesplugin-marketplace.mddoc2,437 bytesprd-intake.mddoc3,311 bytesrai-charter.mddoc2,941 bytesrai-policy.mddoc4,057 bytesralph-charter.mddoc2,236 bytesralph-circuit-breaker.mddoc10,439 bytesralph-instructions.mddoc3,445 bytesralph-reference.mddoc8,828 bytesralph-triage.jsruns16,280 bytesraw-agent-output.mddoc1,336 bytesroster.mddoc2,389 bytesrouting.mddoc1,789 bytesrun-output.mddoc1,500 bytesschedule.jsonasset390 bytesscribe-charter.mddoc3,001 bytesscripts/notes/fetch.ps1runs3,977 bytesscripts/notes/write-note.ps1runs5,449 bytessession-init-reference.mddoc7,194 bytesskill.mdcore587 bytesspawn-reference.mddoc9,351 bytessquad.agent.md.templateasset80,210 bytesworkflow-wiring-appendix-a-code-reviewer.mddoc6,192 bytesworkflow-wiring-appendix-b-documenter.mddoc6,209 bytesworkflow-wiring-guide.mddoc14,233 bytesworkflows/squad-ci.ymlasset506 bytesworkflows/squad-docs.ymlasset1,264 bytesworkflows/squad-heartbeat.ymlasset4,857 bytesworkflows/squad-insider-release.ymlasset2,307 bytesworkflows/squad-issue-assign.ymlasset9,838 bytesworkflows/squad-label-enforce.ymlasset6,935 bytesworkflows/squad-preview.ymlasset1,842 bytesworkflows/squad-promote.ymlasset4,363 bytesworkflows/squad-release.ymlasset2,722 bytesworkflows/squad-triage.ymlasset14,752 bytesworkflows/sync-squad-labels.ymlasset9,408 bytesworktree-reference.mddoc7,553 bytes
Read SKILL.md
## Context
{When and why this skill applies}
## Patterns
{Specific patterns, conventions, or approaches}
## Examples
{Code examples or references}
## Anti-Patterns
{What to avoid}agent-collaboration
When your agent uses it
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
440 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore2,010 bytes
Read SKILL.md
## Context
Every agent on the team follows identical collaboration patterns for worktree awareness, decision recording, and cross-agent communication. These were previously duplicated in every charter's Collaboration section (~300 bytes × 18 agents = ~5.4KB of redundant context). Now centralized here.
The coordinator's spawn prompt already instructs agents to read decisions.md and their history.md. This skill adds the patterns for WRITING decisions and requesting help.
## Patterns
### Worktree Awareness
Use the `TEAM ROOT` path provided in your spawn prompt. All `.squad/` paths are relative to this root. If TEAM ROOT is not provided (rare), run `git rev-parse --show-toplevel` as fallback. Never assume CWD is the repo root.
### Decision Recording
After making a decision that affects other team members, write it to:
`.squad/decisions/inbox/{your-name}-{brief-slug}.md`agent-conduct
When your agent uses it
Shared hard rules enforced across all squad agents
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
283 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore1,345 bytes
Read SKILL.md
## Context
Every squad agent must follow these two hard rules. They were previously duplicated in every charter. Now they live here as a shared skill, loaded once.
## Patterns
### Product Isolation Rule (hard rule)
Tests, CI workflows, and product code must NEVER depend on specific agent names from any particular squad. "Our squad" must not impact "the squad." No hardcoded references to cast-specific agent names in test assertions, CI configs, or product logic. Use generic/parameterized values. If a test needs agent names, use obviously-fake test fixtures (e.g., "test-agent-1", "TestBot").
### Peer Quality Check (hard rule)
Before finishing work, verify your changes don't break existing tests. Run the test suite for files you touched. If CI has been failing, check your changes aren't contributing to the problem. When you learn from mistakes, update your history.md.
## Anti-Patterns
- Don't hardcode dev team agent names in product code or testsarchitectural-proposals
When your agent uses it
How to write comprehensive architectural proposals that drive alignment before code is written
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
1,427 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,335 bytes
Read SKILL.md
## Context
Proposals create alignment before code is written. Cheaper to change a doc than refactor code. Use this pattern when:
- Architecture shifts invalidate existing assumptions
- Product direction changes require new foundation
- Multiple waves/milestones will be affected by a decision
- External dependencies (Copilot CLI, SDK APIs) change
## Patterns
### Proposal Structure (docs/proposals/)
**Required sections:**
1. **Problem Statement** — Why current state is broken (specific, measurable evidence)ci-validation-gates
When your agent uses it
Defensive CI/CD patterns: semver validation, token checks, retry logic, and draft detection
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,072 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Known Failure Modes
- Anti-Patterns
Scripts executed separately
../../../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,562 bytes../../../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
CI workflows must be defensive. These patterns capture lessons from prior release incidents;
they are maintained as version-agnostic gates rather than claims about a particular repository
release.
## Patterns
### Semver Validation Gate
Every publish workflow MUST validate version format before `npm publish`. 4-part versions (e.g., 0.8.21.4) are NOT valid semver — npm mangles them.
```yaml
- name: Validate semver
run: |When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
491 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Checklist — Adding a New CLI Command
- Wiring Patterns by Command Type
- Common Import Pattern
- History
Files
SKILL.mdcore1,844 bytes
Read SKILL.md
# Skill: CLI Command Wiring
**Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.
## Checklist — Adding a New CLI Command
1. **Create command file** in `packages/squad-cli/src/cli/commands/<name>.ts`
- Export a `run<Name>(cwd, options)` async function (or class with static methods for utility modules)
2. **Add routing block** in `packages/squad-cli/src/cli-entry.ts` inside `main()`:
```ts
if (cmd === '<name>') {
const { run<Name> } = await import('./cli/commands/<name>.js');
// parse args, call function
await run<Name>(process.cwd(), options);client-compatibility
When your agent uses it
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,289 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,719 bytes
Read SKILL.md
## Context
Squad runs on multiple Copilot surfaces (CLI, VS Code, JetBrains, GitHub.com). The coordinator must detect its platform and adapt spawning behavior accordingly. Different tools are available on different platforms, requiring conditional logic for agent spawning, SQL usage, and response timing.
## Patterns
### Platform Detection
Before spawning agents, determine the platform by checking available tools:
1. **CLI mode** — `task` tool is available → full spawning control. Use `task` with `agent_type`, `mode`, `model`, `description`, `prompt` parameters. Collect results via `read_agent`.
2. **VS Code mode** — `runSubagent` or `agent` tool is available → conditional behavior. Use `runSubagent` with the task prompt. Drop `agent_type`, `mode`, and `model` parameters. Multiple subagents in one turn run concurrently (equivalent to background mode). Results return automatically — no `read_agent` needed.
coordinator-init-mode
When your agent uses it
The complete two-phase Init Mode protocol the Squad coordinator runs when no team exists yet in the current repo. Phase 1 = propose the team (no files created, wait for user confirm). Phase 2 = create .squad/ scaffolding, casting state, .gitattributes for merge drivers, and the always-on built-ins (Scribe, Ralph, Rai, Fact Checker). Loaded on demand when the coordinator detects no .squad/team.md exists.
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
1,937 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Phase 1: Propose the Team
- Phase 2: Create the Team
Files
SKILL.mdcore7,868 bytes
Read SKILL.md
> **Load this skill when:** you detect that no `.squad/team.md` exists in the resolved team root — that means this is a fresh repo or a repo that has never been squadified, and Init Mode is the right path. The short stub in `squad.agent.md` tells you to load this skill; the full two-phase protocol lives here.
>
> **⚠<U+FE0F> Eager-execution exception:** Init Mode is the ONE exception to the eager-execution / parallel-fan-out doctrine. Phase 1 MUST end with a user confirmation before any file is created. Do not bypass.
## Phase 1: Propose the Team
No team exists yet. **Propose one — but DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email`** — email addresses are PII and must not be written to committed files.
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see the canonical Casting reference at `.squad/templates/casting-reference.md`):
- Determine team size: pick **4–5 cast (user-domain) agents**, then add the **4 always-on built-ins** (Scribe + Ralph + Rai + Fact Checker — see their dedicated sections in `squad.agent.md`). A typical fresh squad has **8–9 total roster entries**, not 4–5.
- Determine assignment shape from the user's project description.
- Derive resonance signals from the session and repo context.coordinator-response-mode
When your agent uses it
Selecting WHO handles work is the Routing table; selecting HOW they handle it (Direct, Lightweight, Standard, Full) is Response Mode. This skill contains the complete decision table, exemplar prompts for each mode, the Lightweight spawn template, and the upgrade rules. Squad coordinator loads this on demand once routing has identified the agent — to pick the right ceremony level for the task.
What it loads into context
Frontmatter
139 tokensAlways available to the agent
SKILL.md
1,172 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Response Mode Selection
- Direct Mode exemplars
- Lightweight Mode exemplars
- Standard Mode exemplars
- Full Mode exemplars
- Mode upgrade rules
- Lightweight Spawn Template
Files
SKILL.mdcore5,018 bytes
Read SKILL.md
> **Load this skill when:** you have routed work to an agent and need to pick the response mode (Direct / Lightweight / Standard / Full). The 1-line stub in `squad.agent.md` is for awareness; this skill is the full decision table + templates.
## Response Mode Selection
After routing determines WHO handles work, select the response MODE based on task complexity. **Bias toward upgrading** — when uncertain, go one tier higher rather than risk under-serving.
| Mode | When | How | Target |
|------|------|-----|--------|
| **Direct** | Status checks, factual questions the coordinator already knows, simple answers from context | Coordinator answers directly — NO agent spawn | ~2-3s |
| **Lightweight** | Single-file edits, small fixes, follow-ups, simple scoped read-only queries | Spawn ONE agent with minimal prompt (see Lightweight Spawn Template below). Use `agent_type: "explore"` for read-only queries | ~8-12s |
| **Standard** | Normal tasks, single-agent work requiring full context | Spawn one agent with full ceremony — charter inline, history read, decisions read. This is the current default | ~25-35s |
| **Full** | Multi-agent work, complex tasks touching 3+ concerns, "Team" requests | Parallel fan-out, full ceremony, Scribe included | ~40-60s |
## Direct Mode exemplarscoordinator-source-of-truth
When your agent uses it
The complete file-by-file source-of-truth hierarchy for Squad: which files are authoritative, which are derived/append-only, who may write each one, who may read each one, and the precedence rules when they conflict. Squad coordinator loads this on demand when it needs to resolve a write conflict, decide where a piece of state belongs, or answer a 'who owns this file' question.
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
1,380 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- State backend note
- File hierarchy
- Rules
Files
SKILL.mdcore5,941 bytesSKILL.mdcore5,941 bytes
Read SKILL.md
> **Load this skill when:** the coordinator (or any agent) needs to resolve a "where does this state belong?" or "who is allowed to write this file?" question — e.g., when about to write `.squad/decisions.md`, when reviewing whether an agent broke the append-only rule, when answering a user question about Squad's file layout, or when adjudicating a conflict between two files. The short summary in `squad.agent.md` is for routing; this skill is the full reference.
## State backend note
Files below marked as **"Derived / append-only"** are **mutable state** — agents access them with runtime state tools (`squad_state_read`, `squad_state_write`, `squad_state_append`, `squad_state_create_if_absent`, `squad_state_delete`, `squad_state_list`). The runtime decides whether the configured backend stores them on disk, git-native state, or an external provider. Files marked as **"Authoritative"** are **static config** and always live on disk regardless of backend.
## File hierarchy
| File | Status | Who May Write | Who May Read |
|------|--------|---------------|--------------|
| `.github/agents/squad.agent.md` | **Authoritative governance.** All roles, handoffs, gates, and enforcement rules. | Repo maintainer (human) | Squad (Coordinator) |
| `.squad/decisions.md` | **Authoritative decision ledger.** Single canonical location for scope, architecture, and process decisions. | Squad (Coordinator) — append only | All agents |
| `.squad/team.md` | **Authoritative roster.** Current team composition. | Squad (Coordinator) | All agents |
| `.squad/routing.md` | **Authoritative routing.** Work assignment rules. | Squad (Coordinator) | Squad (Coordinator) |cross-machine-coordination
When your agent uses it
Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002).
What it loads into context
Frontmatter
58 tokensAlways available to the agent
SKILL.md
3,211 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Usage
- File Formats
- Security Model
- Configuration
- Examples
- Error Handling
- Monitoring & Debugging
- Integration with Ralph Watch
- Migration from Manual Handoff
- Future Enhancements
- Questions?
Files
SKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytes
Read SKILL.md
# Skill: Cross-Machine Coordination Pattern
**Skill ID:** `cross-machine-coordination`
**Owner:** Ralph (Work Monitor)
**Squad Integration:** All agents
**Status:** Design proposal — NOT implemented
> ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
> the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
> and `squad watch` do not poll task files, run whitelisted commands, or
> write result files. Everything below describes a proposed design, not
> current behavior. Do not tell users it works, and do not create task files
> expecting another machine to pick them up.
>cross-squad
When your agent uses it
Coordinating work across multiple Squad instances — discovery, delegation, and disambiguation when the user says 'squad' (the product) vs casual English 'group of agents'.
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
2,402 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Disambiguation: "squad" vs ad-hoc agents
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytes
Read SKILL.md
## Context
> **Read this FIRST any time the user says "squad" as a thing to spawn, delegate to, address, or fan out to** — e.g., *"spawn two squads of designers and devs"*, *"ask the other squad"*, *"delegate to a squad"*. In Squad-PRODUCT vocabulary, "squad" is a **peer** (an independent installation with its own `.squad/`, `team.md`, MCP server, and agents) — NOT a generic English synonym for "team" or "group". Do not fan out raw `task` agents inside your own coordinator context when the user means "another squad". Use the discovery and communication patterns below (and the companion `cross-squad-communication` skill for the actual protocols).
When an organization runs multiple Squad instances (e.g., platform-squad, frontend-squad, data-squad), those squads need to discover each other, share context, and hand off work across repository boundaries. This skill teaches agents how to coordinate across squads without creating tight coupling.
> **Companion skill — for protocol details:** `cross-squad-communication/SKILL.md` covers the four communication patterns (synchronous CLI, read-only knowledge query, git-based async, and GitHub-issue-based delegation) once a peer squad is discovered via the registry below. This skill answers "who?" — the companion answers "how?".
Cross-squad orchestration applies when:
- A task requires capabilities owned by another squad
- An architectural decision affects multiple squads
- A feature spans multiple repositories with different squads
- A squad needs to request infrastructure, tooling, or support from another squad
cross-squad-communication
When your agent uses it
Protocol for sending queries, delegating tasks, and sharing context between independent Squad instances across different repositories
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
4,739 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Universal rule: every `copilot` spawn into a peer squad MUST pass `--agent squad`
- Discovery Protocol
- Platform Compatibility Matrix
- Examples
- Anti-Patterns
- Validation Status
Files
SKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytes
Read SKILL.md
## Context
When multiple repositories each have their own Squad (AI team), they need to exchange information: knowledge queries, PR reviews, task delegation, and dependency analysis. Each squad has its own agents, MCP tools, and issue tracker — there is no shared runtime.
> **Companion skill — read first:** `cross-squad/SKILL.md` covers **discovery** of peer squads via `squad registry add/list/remove`. This skill picks up after a peer is known and covers the **communication protocols** themselves — the four numbered patterns below: Pattern 0 (synchronous CLI), Pattern 1 (read-only knowledge query), Pattern 2 (git-based async), and Pattern 3 (GitHub-issue-based delegation). A separate non-numbered appendix (Cross-Repo Dependency Scan) is provided as a related analysis tool, not a communication pattern. The two skills are designed to be used together.
**When this skill applies:**
- A squad agent needs information from another squad-enabled repo
- A task needs to be delegated to another squad
- Cross-repo dependency analysis is needed
- PR review requests span repo boundaries
**Key constraint:** Each squad has its own runtime, MCP tools, and issue tracker. Cross-squad communication can be **synchronous** (via CLI session targeting the other repo) or **asynchronous** (file-based or issue-based). The coordinator decides which approach fits.
distributed-mesh
When your agent uses it
How to coordinate with squads on different machines using git as transport
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
3,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Patterns
- Examples
- AGENT WORKFLOW (Deterministic Setup)
- Anti-Patterns
Scripts executed separately
sync-mesh.ps1sync-mesh.sh
Files
SKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesmesh.json.exampleasset762 bytessync-mesh.ps1runs3,861 bytessync-mesh.shruns3,395 bytes
Read SKILL.md
## SCOPE
**✅ THIS SKILL PRODUCES (exactly these, nothing more):**
1. **`mesh.json`** — Generated from user answers about zones and squads (which squads participate, what zone each is in, paths/URLs for each), using `mesh.json.example` in this skill's directory as the schema template
2. **`sync-mesh.sh` and `sync-mesh.ps1`** — Copied from this skill's directory into the project root (these are bundled resources, NOT generated code)
3. **Zone 2 state repo initialization** (if applicable) — If the user specified a Zone 2 shared state repo, run `sync-mesh.sh --init` to scaffold the state repo structure
4. **A decision entry** in `.squad/decisions/inbox/` documenting the mesh configuration for team awareness
**❌ THIS SKILL DOES NOT PRODUCE:**
- **No application code** — No validators, libraries, or modules of any kind
- **No test files** — No test suites, test cases, or test scaffolding
- **No GENERATING sync scripts** — They are bundled with this skill as pre-built resources. COPY them, don't generate them.docs-standards
When your agent uses it
Microsoft Style Guide + Squad-specific documentation patterns
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
524 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore2,270 bytesSKILL.mdcore2,270 bytes
Read SKILL.md
## Context
Squad documentation follows the Microsoft Style Guide with Squad-specific conventions. Consistency across docs builds trust and improves discoverability.
## Patterns
### Microsoft Style Guide Rules
- **Sentence-case headings:** "Getting started" not "Getting Started"
- **Active voice:** "Run the command" not "The command should be run"
- **Second person:** "You can configure..." not "Users can configure..."
- **Present tense:** "The system routes..." not "The system will route..."
- **No ampersands in prose:** "and" not "&" (except in code, brand names, or UI elements)
### Squad Formatting Patternse2e-template-testing
When your agent uses it
End-to-end validation of coordinator and agent template changes
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
6,164 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- When To Use
- Prerequisites
- Workflow
- Progress Reporting
- Test Matrix Template
- Tips
- Fast-Fail Rules
- PII Protection — Mandatory
- Anti-Patterns
- Sandbox / Permission Notes
- Confidence
Files
SKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytes
Read SKILL.md
## Context
Squad's coordinator prompt (`squad.agent.md`) and agent charters (e.g.
`scribe-charter.md`) are shipped as templates in `.squad-templates/`. Changes to
these files affect how every squad session behaves — but unit tests can't catch
prompt-level regressions because the prompts are interpreted by an LLM at
runtime.
This skill describes how to validate template changes end-to-end by running real
squad sessions against a locally-built CLI that includes your modified templates.
## When To Use
- You changed `.squad-templates/squad.agent.md` (coordinator prompt)economy-mode
When your agent uses it
Shifts Layer 3 model selection to cost-optimized alternatives when economy mode is active.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,396 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Activation Methods
- Economy Model Selection Table
- AGENT WORKFLOW
- Config Schema
- Anti-Patterns
Files
SKILL.mdcore5,547 bytes
Read SKILL.md
## SCOPE
✅ THIS SKILL PRODUCES:
- A modified Layer 3 model selection table applied when economy mode is active
- `economyMode: true` written to `.squad/config.json` when activated persistently
- Spawn acknowledgments with `💰` indicator when economy mode is active
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Cost reports or billing artifacts
- Changes to Layer 0, Layer 1, or Layer 2 resolution (user intent always wins)
## Context
error-recovery
When your agent uses it
Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
785 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Retry with Backoff
- 2. Fallback Alternatives
- 3. Diagnose-and-Fix
- 4. Escalate with Context
- 5. Graceful Degradation
- Applying These Patterns
Files
SKILL.mdcore3,466 bytesSKILL.mdcore3,466 bytes
Read SKILL.md
# Error Recovery Patterns
Standard recovery patterns for all squad agents. When something fails, **adapt** — don't just report the failure.
---
## 1. Retry with Backoff
**When:** Transient failures — API timeouts, rate limits, network errors, temporary service unavailability.
**Pattern:**
1. Wait briefly, then retry (start at 2s, double each attempt)
2. Maximum 3 retries before escalating
3. Log each attempt with the error receivedexternal-comms
When your agent uses it
Workflow for scanning, drafting, and presenting community responses with a human review gate
What it loads into context
Frontmatter
228 tokensAlways available to the agent
SKILL.md
3,012 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore13,159 bytesSKILL.mdcore13,159 bytes
Read SKILL.md
## Context
Phase 1 is **draft-only mode**.
- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.
- **Human review gate is mandatory** — the communications owner never posts autonomously.
- Every action is logged to `.squad/comms/audit/`.
- This workflow is triggered manually only ("check community") — no automated activation in Phase 1.
## Patterns
### 1. Scan
Find unanswered community items with GitHub MCP tools first, or `gh issue list` / `gh api` as fallback for issues and discussions.fact-checking
When your agent uses it
Review and validate claims using counter-hypothesis testing. Use when verifying technical content, checking references, validating API endpoints, or performing quality assurance on deliverables.
What it loads into context
Frontmatter
97 tokensAlways available to the agent
SKILL.md
469 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore2,046 bytes
Read SKILL.md
# Skill: Fact Checking
## Context
Codifies the challenger agent review output format and methodology so any agent performing fact-checking or review produces consistent, structured output.
## Pattern
### Review Methodology
For every claim or deliverable under review:
1. Ask: "What evidence supports this? What would disprove it?"
2. Generate counter-hypotheses and test them against available data
3. Verify URLs, package names, API endpoints, and external references actually exist
4. Flag confidence levels: ✅ Verified, ⚠<U+FE0F> Unverified, ❌ Contradictedgh-auth-isolation
When your agent uses it
Safely manage multiple GitHub identities (EMU + personal) in agent workflows
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
1,543 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytesSKILL.mdcore6,624 bytes
Read SKILL.md
## Context
Many developers use GitHub through an Enterprise Managed User (EMU) account at work while maintaining a personal GitHub account for open-source contributions. AI agents spawned by Squad inherit the shell's default `gh` authentication — which is usually the EMU account. This causes failures when agents try to push to personal repos, create PRs on forks, or interact with resources outside the enterprise org.
This skill teaches agents how to detect the active identity, switch contexts safely, and avoid mixing credentials across operations.
## Patterns
### Detect Current Identity
Before any GitHub operation, check which account is active:
```bash
gh auth statusgh-aw-enlistment
When your agent uses it
Enlist a repository into Squad by installing the supported GitHub Agentic Workflows (gh-aw) bootstrap — with strict compilation, an explicit safe-update allowlist, and a human-reviewed bootstrap PR. Never auto-merges.
What it loads into context
Frontmatter
308 tokensAlways available to the agent
SKILL.md
7,239 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytes
Read SKILL.md
## Context
Use this skill when a user wants to **enlist a repository into Squad** through
**GitHub Agentic Workflows (`gh aw`)** — e.g. "set up Squad agentic workflows",
"enlist this repo in Squad", or "install Squad gh-aw workflows". The end state is
the `/squad` slash command being live on the repo, delivered as a **human-reviewed
bootstrap pull request** — never an auto-merge.
This skill **operationalizes** the supported bootstrap path. It does not summarize
it: each step below is a gate with explicit success evidence and **STOP conditions**.
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this
skill ever disagree, the guide wins — re-read it before proceeding.
git-workflow
When your agent uses it
Squad branching model: dev-first workflow with insiders preview channel
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,581 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Branch Naming Convention
- Workflow for Issue Work
- Parallel Multi-Issue Work (Worktrees)
- Multi-Repo Downstream Scenarios
- Anti-Patterns
- Promotion Pipeline
Files
SKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytes
Read SKILL.md
## Context
Squad uses a three-branch model. **All feature work starts from `dev`, not `main`.**
| Branch | Purpose | Publishes |
|--------|---------|-----------|
| `main` | Released, tagged, in-npm code only | `npm publish` on tag |
| `dev` | Integration branch — all feature work lands here | `npm publish --tag preview` on merge |
| `insiders` | Early-access channel — synced from dev | `npm publish --tag insiders` on sync |
## Branch Naming Convention
Issue branches MUST use: `squad/{issue-number}-{kebab-case-slug}`
github-multi-account
When your agent uses it
Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
What it loads into context
Frontmatter
80 tokensAlways available to the agent
SKILL.md
1,038 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Activate
- What to Do (as the AI agent)
- After Setup — Usage Rules
- Repo-Specific Account Binding
- For Squad Agents
Files
SKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytes
Read SKILL.md
# GitHub Multi-Account — AI-Driven Setup
## When to Activate
When the user has multiple GitHub accounts (check with `gh auth status`). If you see 2+ accounts listed, this skill applies.
## What to Do (as the AI agent)
### Step 1: Detect accounts
Run: `gh auth status`
Look for multiple accounts. Note which usernames are listed.
### Step 2: Ask the user
Ask: "I see you have multiple GitHub accounts: {list them}. Which one is your personal account and which is your work/EMU account?"
history-hygiene
When your agent uses it
Record reconciled final outcomes to history.md, not intermediate requests or obsolete claims
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
545 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Patterns
- Examples
- Stop conditions
- Anti-Patterns
Files
SKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytes
Read SKILL.md
## Context
History files are cold-start context, not the source of truth for live behavior. Stale or incorrect
entries poison decision-making downstream. A stale-history incident proved that an old history claim
can cause future spawns to repeat a reversed decision. Reconcile an operational claim with current
implementation, workflows, tests, documentation, skills, and applicable decisions before acting on it.
## Scope
- Activate when recording, correcting, or promoting finalized operational history.
- Allowed targets: the explicitly selected history file and a focused reusable skill when the rule
applies beyond one role.
- Exclusions: intermediate requests, stale assignments, roster metadata, and unrelated cleanup.
humanizer
When your agent uses it
Tone enforcement patterns for external-facing community responses
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,014 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,430 bytes
Read SKILL.md
## Context
Use this skill whenever the communications owner drafts external-facing responses for issues or discussions.
- Tone must be warm, helpful, and human-sounding — never robotic or corporate.
- The repository's communication policy applies everywhere: **Humanized tone is mandatory**.
- This applies to **all external-facing content** drafted by the communications owner in Phase 1 issues/discussions workflows.
## Patterns
1. **Warm opening** — Start with acknowledgment ("Thanks for reporting this", "Great question!")
2. **Active voice** — "We're looking into this" not "This is being investigated"
3. **Second person** — Address the person directly ("you" not "the user")
4. **Conversational connectors** — "That said...", "Here's what we found...", "Quick note:"init-mode
When your agent uses it
Team initialization flow (Phase 1 proposal + Phase 2 creation)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
1,749 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,100 bytesSKILL.mdcore7,100 bytes
Read SKILL.md
## Context
Init Mode activates when `.squad/team.md` does not exist, or exists but has zero roster entries under `## Members`. The coordinator proposes a team (Phase 1), waits for user confirmation, then creates the team structure (Phase 2).
## Patterns
### Phase 1: Propose the Team
No team exists yet. Propose one — but **DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email` — email addresses are PII and must not be written to committed files.**
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see that section):
- Determine team size (typically 4–5 + Scribe).iterative-retrieval
When your agent uses it
Max-3-cycle protocol for agent sub-tasks with WHY context and coordinator validation. Use when spawning sub-agents to complete scoped work.
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,385 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Spawn Prompt Template
- 3-Cycle Protocol
- Coordinator Validation Checklist
- When to Escalate vs Retry
- Issue Dedup Check (Mandatory)
- Mandatory Output Requirement (Research-Then-Execute)
- Anti-Patterns
- Examples
Files
SKILL.mdcore6,341 bytesSKILL.mdcore6,341 bytes
Read SKILL.md
# Iterative Retrieval Skill
Squad agents frequently spawn sub-agents to complete scoped work. Without structure, these
handoffs become vague, cycles multiply, and outputs land without being checked. The
**Iterative Retrieval Pattern** caps cycles at 3, mandates WHY context in every spawn, and
requires the coordinator to validate agent output before closing an issue.
---
## Spawn Prompt Template
Every agent spawn must include the following four sections. Copy and fill in the template:
```model-selection
When your agent uses it
Determines which LLM model to use for each agent spawn
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- 5-Layer Model Resolution Hierarchy
- AGENT WORKFLOW
- Config Schema
- Fallback Chains
Files
SKILL.mdcore8,082 bytes
Read SKILL.md
# Model Selection
> Determines which LLM model to use for each agent spawn.
## SCOPE
✅ THIS SKILL PRODUCES:
- A resolved `model` parameter for every `task` tool call
- Persistent model preferences in `.squad/config.json`
- Spawn acknowledgments that include the resolved model
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Model performance benchmarksnap
When your agent uses it
Context hygiene — compress, prune, archive .squad/ state
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
208 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What It Does
- When To Use
- Invocation
- Confidence
Files
SKILL.mdcore792 bytes
Read SKILL.md
# Skill: nap
> Context hygiene — compress, prune, archive .squad/ state
## What It Does
Reclaims context window budget by compressing agent histories, pruning old logs,
archiving stale decisions, and cleaning orphaned inbox files.
## When To Use
- Before heavy fan-out work (many agents will spawn)
- When history.md files exceed 15KB
- When .squad/ total size exceeds 1MBnotification-routing
When your agent uses it
Route agent notifications to specific channels by type — prevent alert fatigue from single-channel flooding
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
730 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
- Distributed Systems Pattern
Files
SKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytes
Read SKILL.md
## Context
When a Squad grows beyond a few agents, notifications flood a single channel — failure alerts drown in daily
briefings, tech news buries security findings, and everything gets ignored. This is the pub-sub problem:
a single message queue for everything is a recipe for missed alerts.
The fix is **topic-based routing**: agents tag notifications with a channel type, and a routing function
sends them to the appropriate destination.
**Trigger symptoms:**
- Important alerts missed because they're buried in routine notifications
- Team members turning off notifications entirely (signal overwhelm)
- Onboarding friction: "where do I look for X?"
pr-review-response
When your agent uses it
Teaches agents to reply to PR review comment threads after fixing issues, making resolutions traceable
What it loads into context
Frontmatter
154 tokensAlways available to the agent
SKILL.md
2,574 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- SCOPE
- Patterns
- AGENT WORKFLOW (Summary)
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytes
Read SKILL.md
## Context
When an agent fixes code in response to PR review comments (from Copilot, a human reviewer, or any GitHub reviewer), the fix alone is not enough. The reviewer needs to see — on the PR thread itself — which comments were addressed and how. Without replies, comments stay visually unresolved, reviewers must re-read the entire diff to verify fixes, and there's no traceable link between feedback and resolution.
Use this skill whenever:
- You are fixing code based on PR review feedback
- You are addressing Copilot review suggestions
- You are responding to reviewer-requested changes on a PR
- A squad member hands you review comments to resolve
## SCOPE
✅ THIS SKILL PRODUCES:
- Reply comments on each review thread explaining the fixpr-screenshots
When your agent uses it
Capture Playwright screenshots and embed them in GitHub PR descriptions
What it loads into context
Frontmatter
62 tokensAlways available to the agent
SKILL.md
1,065 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../../../scripts/capture-pr-screenshots.mjs
Files
SKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytesSKILL.mdcore4,604 bytes../../../../../scripts/capture-pr-screenshots.mjsruns1,855 bytes
Read SKILL.md
## Context
When a PR includes visual changes (docs sites, UI components, generated pages), reviewers
need to see what the PR delivers without checking out the branch. Screenshots belong in
the **PR description body**, not as committed files and not as text descriptions.
Use this skill whenever:
- A PR touches docs site pages (Astro, Starlight, etc.)
- A PR adds or changes UI components
- A PR generates visual artifacts (TypeDoc, Storybook, diagrams)
- Playwright tests already capture screenshots as part of testing
## Patterns
project-conventions
When your agent uses it
Core conventions and patterns for this codebase
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,555 bytes
Read SKILL.md
## Context
> **This is a starter template.** Replace the placeholder patterns below with your actual project conventions. Skills train agents on codebase-specific practices — accurate documentation here improves agent output quality.
## Patterns
### [Pattern Name]
Describe a key convention or practice used in this codebase. Be specific about what to do and why.
### Error Handling
<!-- Example: How does your project handle errors? -->
<!-- - Use try/catch with specific error types? -->ralph-two-pass-scan
When your agent uses it
Cuts GitHub API calls by separating lightweight list scanning from full hydration
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
345 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore1,336 bytes
Read SKILL.md
# Skill: Ralph — Two-Pass Issue Scanning
**Confidence:** high
**Domain:** work-monitoring
**Last validated:** 2026-03-24
## Context
Cuts GitHub API calls from N+1 to ~7 per round (~72% reduction) by separating list scanning from full hydration.
Addresses the scanning inefficiency described in issue #596.
## Pattern
### Pass 1 — Lightweight Scan
```reflect
When your agent uses it
Learning capture system that extracts HIGH/MED/LOW confidence patterns from conversations to prevent repeating mistakes. Use after user corrections ("no", "wrong"), praise ("perfect", "exactly"), or when discovering edge cases. Complements .squad/agents/{agent}/history.md and .squad/decisions.md.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Integration with Squad Architecture
- Triggers
- Process
- Usage Examples
- When to Use
- See Also
Files
SKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytes
Read SKILL.md
# Reflect Skill
**Critical learning capture system** for Squad. Prevents repeating mistakes and preserves successful patterns across sessions.
Analyze conversations and propose improvements to squad knowledge based on what worked, what didn't, and edge cases discovered. **Every correction is a learning opportunity.**
---
## Integration with Squad Architecture
**Reflect complements existing Squad knowledge systems:**
1. **Governed runtime memory** — Durable observations and decisions persisted through configured state tools
2. **Tracked governing source** — Coordinator prompts, skills, and other behavior-defining files changed through focused pull requestsrelease-process
When your agent uses it
Prepare, publish, recover, and verify Squad insider, preview, and stable releases
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
2,578 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Non-negotiable rules
- Required credentials
- Human release-trigger boundary
- Prepare a release
- Publish a preview release
- Publish an insider release
- Publish a stable release
- Verify publication
- Recovery
- After a stable release
Files
SKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytes
Read SKILL.md
# Release Process
This is the canonical Squad release runbook. Squad has three release channels
and two long-lived branches:
| Source | Version | GitHub | npm | Standalone | Homebrew/WinGet |
|--------|---------|--------|-----|------------|-----------------|
| Insider dispatch from `dev` | Generated `X.Y.Z-insider.N` | Prerelease | `insider` | Yes | Yes |
| Release dispatch from `dev` | `X.Y.Z-preview.N` | Prerelease | `preview` | Yes | Yes |
| Promotion PR merged to `main` (or legacy workflow push) | `X.Y.Z` | Stable/latest | `latest` | Yes | Yes |
The repository does not use a staging `preview` branch. Preview is a release
channel, not a branch.
reskill
When your agent uses it
Extract durable operating knowledge before slimming or replacing team charters
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,223 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Process
- Patterns
- Anti-Patterns
Files
SKILL.mdcore5,760 bytesSKILL.mdcore5,760 bytes
Read SKILL.md
## Context
When the coordinator hears "team, reskill" (or similar: "optimize context", "slim down charters"), trigger a team-wide optimization pass. The goal: reduce per-agent context consumption by extracting shared patterns from charters and histories into reusable skills.
This is a periodic maintenance activity. Run whenever charter/history bloat is suspected or a
team will be re-cast. Charters and histories are evidence, not the authority for current product
behavior: reconcile candidates against current source, workflows, tests, documentation, skills,
and decisions before preserving them.
## Process
### Step 1: Audit
Read all active agent charters and histories, then inventory existing skills. Measure byte sizes.
Identify:When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
1,352 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Purpose
- Core Function: Test-RetroOverdue
- Coordinator Integration
- Action Item Enforcement
- Cadence Enforcement
- Skill Metadata
Files
SKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytes
Read SKILL.md
# Skill: Retro Enforcement
## Purpose
Ensure retrospectives happen on schedule and that their action items are tracked in GitHub Issues — not markdown checklists.
This skill addresses a specific, measured failure mode: **0% completion rate on markdown retro action items across 6 consecutive retrospectives**. GitHub Issues have an 85%+ completion rate in the same squad. The format was the problem, not the people.
## Core Function: Test-RetroOverdue
```powershell
function Test-RetroOverdue {
param(
[string]$LogDir = ".squad/log",
[int]$WindowDays = 7,reviewer-protocol
When your agent uses it
Reviewer rejection workflow and strict lockout semantics
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
1,009 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,658 bytes
Read SKILL.md
## Context
When a team member has a **Reviewer** role (e.g., Tester, Code Reviewer, Lead), they may approve or reject work from other agents. On rejection, the coordinator enforces strict lockout rules to ensure the original author does NOT self-revise. This prevents defensive feedback loops and ensures independent review.
## Patterns
### Reviewer Rejection Protocol
Quality and test reviewers may block a PR or merge on quality, coverage, or regression grounds. Treat that block as a rejection that enters the protocol below.
When a team member has a **Reviewer** role:
- Reviewers may **approve** or **reject** work from other agents.
- On **rejection**, the Reviewer may choose ONE of:secret-handling
When your agent uses it
Never read .env files or write secrets to .squad/ committed files
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
1,994 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,524 bytesSKILL.mdcore7,524 bytes
Read SKILL.md
## Context
Spawned agents have read access to the entire repository, including `.env` files containing live credentials. If an agent reads secrets and writes them to `.squad/` files (decisions, logs, history), Scribe auto-commits them to git, exposing them in remote history. This skill codifies absolute prohibitions and safe alternatives.
## Patterns
### Prohibited File Reads
**NEVER read these files:**
- `.env` (production secrets)
- `.env.local` (local dev secrets)
- `.env.production` (production environment)
- `.env.development` (development environment)
- `.env.staging` (staging environment)session-recovery
When your agent uses it
Find and resume interrupted Copilot CLI sessions using session_store queries
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,215 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytes
Read SKILL.md
## Context
Squad agents run in Copilot CLI sessions that can be interrupted — terminal crashes, network drops, machine restarts, or accidental window closes. When this happens, in-progress work may be left in a partially-completed state: branches with uncommitted changes, issues marked in-progress with no active agent, or checkpoints that were never finalized.
Copilot CLI stores session history in a SQLite database called `session_store` (read-only, accessed via the `sql` tool with `database: "session_store"`). This skill teaches agents how to query that store to detect interrupted sessions and resume work.
## Patterns
### 1. Find Recent Sessions
Query the `sessions` table filtered by time window. Include the last checkpoint to understand where the session stopped:
```sql
SELECTsquad
When your agent uses it
Squad's command catalog and interactive menu. Invoke via /squad (slash command) or natural language ("squad commands", "what can squad do", "show me squad options"). Presents categorized operations (Install & Upgrade, Team Management, Issues & PRs, Plugins & Skills, Model & Cost, Sessions & State) as an interactive picker. Routes to the right squad CLI command or the Squad coordinator agent.
What it loads into context
Frontmatter
103 tokensAlways available to the agent
SKILL.md
2,680 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,874 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill (via `/squad` slash command, "squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-commands
When your agent uses it
Categorized catalog of common Squad operations. Coordinator reads this file and presents it as an interactive menu when the user asks for available commands or help.
What it loads into context
Frontmatter
82 tokensAlways available to the agent
SKILL.md
2,652 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,748 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill ("squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-conventions
When your agent uses it
Core conventions and patterns used in the Squad codebase
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
824 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore3,440 bytesSKILL.mdcore3,440 bytes
Read SKILL.md
## Context
These conventions apply to all work on the Squad CLI tool (`create-squad`). Squad is a zero-dependency Node.js package that adds AI agent teams to any project. Understanding these patterns is essential before modifying any Squad source code.
## Patterns
### Zero Dependencies
Squad has zero runtime dependencies. Everything uses Node.js built-ins (`fs`, `path`, `os`, `child_process`). Do not add packages to `dependencies` in `package.json`. This is a hard constraint, not a preference.
### Node.js Built-in Test Runner
Tests use `node:test` and `node:assert/strict` — no test frameworks. Run with `npm test`. Test files live in `test/`. The test command is `node --test test/`.
### Error Handling — `fatal()` Pattern
All user-facing errors use the `fatal(msg)` function which prints a red `✗` prefix and exits with code 1. Never throw unhandled exceptions or print raw stack traces. The global `uncaughtException` handler calls `fatal()` as a safety net.
squad-help
When your agent uses it
How to actually use Squad — Squad is a custom Copilot agent (invoked via the task tool with agent_type='Squad'), not a skill. This file explains the right invocation paths for setting up a team, listing squad commands, and initializing Squad in a new project.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,043 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Squad is a custom agent, not a skill
- How to actually use Squad
- What NOT to do
- How this skill was discovered
- See also
Files
SKILL.mdcore4,198 bytesSKILL.mdcore4,198 bytes
Read SKILL.md
# Skill: squad-help
> **Quick reference.** If you're reading this because a user said "use squad" or "squad" or "set up a squad", you're in the right place — read on for the correct invocation paths.
---
## Squad is a custom agent, not a skill
The Squad framework registers a **custom Copilot CLI agent** at `.github/agents/squad.agent.md`. The agent is named **`Squad`** and its description is *"Your AI team. Describe what you're building, get a team of specialists that live in your repo."*
Copilot CLI agents and skills are different things:
| Thing | How to invoke | Example |
|---|---|---|squad-version-check
When your agent uses it
Internals of how @bradygaster/squad-cli stamps its version, how `squad upgrade` works (what it preserves vs overwrites), and how to probe the npm registry for the latest version from a coordinator prompt.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,702 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What This Skill Covers
- Package & Registry Facts
- Version Stamping Mechanism
- `squad upgrade` Behavior
- `.squad/config.json` — What It Holds
- Version-Check Probe (npm Registry)
- Upstream OS-Specific Cache
- Repo-Local Cache Convention: `.squad/.cache/version-check.json`
- Key File Paths (installed CLI)
Files
SKILL.mdcore6,296 bytes
Read SKILL.md
# SKILL: Squad CLI Internals — Version Stamping & Upgrade Mechanics
**Confidence:** medium
**Discovered by:** repository investigation
**Date:** 2026-05-26
**Validated in:** Issue #1173 recon (bradygaster/squad)
---
## What This Skill Covers
Reusable knowledge about how `@bradygaster/squad-cli` stamps its version into `squad.agent.md`, how `squad upgrade` works, what it preserves vs. overwrites, and how to probe the npm registry for the latest version from a coordinator prompt.
---test-discipline
When your agent uses it
Design assertions that prove the actual contract and fail on realistic regressions
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytes
Read SKILL.md
## Context
Use this skill whenever a change affects an API, public interface, workflow contract, generated
artifact, or user-observable behavior. A passing status-only test is not evidence that the
intended artifact or contract is correct.
## Patterns
- Update tests in the same change as an API, signature, export, or public-interface change.
- Make assertions name and inspect the offending artifact, output, or configuration field—not just
an exit code or broad success status.
- Mutation-test critical gates by changing the real source or emitted artifact and proving the
assertion turns red.
- Exercise every caller path for a changed parser, normalizer, or helper; isolated unit coveragetiered-memory
When your agent uses it
Three-tier agent memory model (hot/cold/wiki) for context reduction per spawn
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
1,775 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Memory Tiers
- When to Load Each Tier
- Spawn Template Pattern
- Integration with Scribe Agent (design — not yet implemented)
- Implementation Checklist (tracked in #1264)
- References
- Spawn Template
- Task
- WHY
- Success Criteria
- Memory Context
- Escalation
- Notes
Files
SKILL.mdcore7,506 bytes
Read SKILL.md
# Skill: Tiered Agent Memory
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, and spawn-template tier-aware reads are tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264). Until those land, agents continue to load full `history.md` + `decisions.md` on every spawn.
## Overview
Squad agents today load their full context history on every spawn, which grows unboundedly across sessions. The Tiered Agent Memory model proposes a three-tier separation so agents only load the bytes that are actually relevant to the current task, with older context kept available on demand.
---
## Memory Tiers
### 🔥 Hot Tier — Current Session Context
- **Size target:** keep small (~2–4KB typical)versioning-policy
When your agent uses it
SemVer rules for Squad's stable, preview, insider, and local package versions
What it loads into context
Frontmatter
60 tokensAlways available to the agent
SKILL.md
1,153 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- 1. Supported versions
- 2. Package versions stay in lockstep
- 3. Prerelease workspace dependency rule
- 4. Local build versions are ephemeral
- 5. Release lifecycle
- 6. Ownership
- 7. CI enforcement
- Quick reference
Scripts executed separately
../../../../../scripts/bump-build.mjs
Files
SKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytes../../../../../scripts/bump-build.mjsruns4,324 bytes
Read SKILL.md
## Context
Squad publishes `@bradygaster/squad-sdk` and
`@bradygaster/squad-cli` from one npm workspace. Their versions and the root
version move together. A mismatched prerelease dependency can make npm silently
resolve an older registry SDK instead of the local workspace.
## 1. Supported versions
| Use | Version | Committed branch |
|---|---|---|
| Stable release | `MAJOR.MINOR.PATCH` | `main`, and briefly `dev` during promotion |
| Preview release | `MAJOR.MINOR.PATCH-preview.N` | `dev` |
| Insider snapshot | `MAJOR.MINOR.PATCH-insider.N` | Generated by the insider workflow |windows-compatibility
When your agent uses it
Cross-platform path handling and command patterns
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
1,572 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../../../scripts/fix-crlf-worktree.mjs
Files
SKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytes../../../../../scripts/fix-crlf-worktree.mjsruns5,444 bytes
Read SKILL.md
## Context
Squad runs on Windows, macOS, and Linux. Several bugs have been traced to platform-specific assumptions: ISO timestamps with colons (illegal on Windows), `git -C` with Windows paths (unreliable), forward-slash paths in Node.js on Windows.
## Patterns
### Filenames & Timestamps
- **Never use colons in filenames:** ISO 8601 format `2026-03-15T05:30:00Z` is illegal on Windows
- **Use `safeTimestamp()` utility:** Replaces colons with hyphens → `2026-03-15T05-30-00Z`
- **Centralize formatting:** Don't inline `.toISOString().replace(/:/g, '-')` — use the utility
### Git Commands
- **Never use `git -C {path}`:** Unreliable with Windows paths (backslashes, spaces, drive letters)
- **Always `cd` first:** Change directory, then run git commands{skill-name}
When your agent uses it
{what this skill teaches agents}
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
145 tokensWhen the skill is activated
Reference documents
48,671 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
package.jsonralph-triage.jsscripts/notes/fetch.ps1scripts/notes/write-note.ps1
Files
README.mddoc7,063 bytesafter-agent-reference.mddoc5,750 bytescasting-history.jsonasset70 bytescasting-policy.jsonasset823 bytescasting-reference.mddoc7,679 bytescasting-registry.jsonasset146 bytescasting/Futurama.jsonasset108 bytesceremonies.mddoc1,882 bytesceremony-reference.mddoc3,162 bytescharter.mddoc2,053 bytesclient-compatibility-reference.mddoc3,758 bytesconstraint-tracking.mddoc1,192 bytescooperative-rate-limiting.mddoc7,222 bytescopilot-agent.mddoc3,777 bytescopilot-instructions.mddoc3,667 bytesfact-checker-charter.mddoc2,621 bytesfact-checker-policy.mddoc6,104 byteshistory.mddoc255 bytesidentity/now.mddoc209 bytesidentity/wisdom.mddoc384 bytesissue-lifecycle.mddoc11,524 byteskeda-scaler.mddoc5,732 bytesloop.mddoc1,378 bytesmachine-capabilities.mddoc2,415 bytesmcp-config.mddoc3,129 bytesmodel-selection-reference.mddoc16,277 bytesmulti-agent-format.mddoc887 bytesnotes-protocol.mddoc5,513 bytesorchestration-log.mddoc1,104 bytespackage.jsonruns28 bytesplugin-marketplace.mddoc2,437 bytesprd-intake.mddoc3,311 bytesrai-charter.mddoc2,941 bytesrai-policy.mddoc4,057 bytesralph-charter.mddoc2,236 bytesralph-circuit-breaker.mddoc10,439 bytesralph-instructions.mddoc3,445 bytesralph-reference.mddoc8,828 bytesralph-triage.jsruns16,280 bytesraw-agent-output.mddoc1,336 bytesroster.mddoc2,389 bytesrouting.mddoc1,789 bytesrun-output.mddoc1,500 bytesscribe-charter.mddoc3,001 bytesscripts/notes/fetch.ps1runs3,977 bytesscripts/notes/write-note.ps1runs5,449 bytessession-init-reference.mddoc7,194 bytesskill.mdcore587 bytesspawn-reference.mddoc9,351 bytessquad.agent.md.templateasset80,210 bytesworkflow-wiring-appendix-a-code-reviewer.mddoc6,192 bytesworkflow-wiring-appendix-b-documenter.mddoc6,209 bytesworkflow-wiring-guide.mddoc14,233 bytesworkflows/squad-ci.ymlasset506 bytesworkflows/squad-docs.ymlasset1,264 bytesworkflows/squad-heartbeat.ymlasset4,857 bytesworkflows/squad-insider-release.ymlasset2,307 bytesworkflows/squad-issue-assign.ymlasset9,838 bytesworkflows/squad-label-enforce.ymlasset6,935 bytesworkflows/squad-preview.ymlasset1,842 bytesworkflows/squad-promote.ymlasset4,363 bytesworkflows/squad-release.ymlasset2,722 bytesworkflows/squad-triage.ymlasset14,752 bytesworkflows/sync-squad-labels.ymlasset9,408 bytesworktree-reference.mddoc7,553 bytes
Read SKILL.md
## Context
{When and why this skill applies}
## Patterns
{Specific patterns, conventions, or approaches}
## Examples
{Code examples or references}
## Anti-Patterns
{What to avoid}agent-collaboration
When your agent uses it
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
440 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore2,010 bytes
Read SKILL.md
## Context
Every agent on the team follows identical collaboration patterns for worktree awareness, decision recording, and cross-agent communication. These were previously duplicated in every charter's Collaboration section (~300 bytes × 18 agents = ~5.4KB of redundant context). Now centralized here.
The coordinator's spawn prompt already instructs agents to read decisions.md and their history.md. This skill adds the patterns for WRITING decisions and requesting help.
## Patterns
### Worktree Awareness
Use the `TEAM ROOT` path provided in your spawn prompt. All `.squad/` paths are relative to this root. If TEAM ROOT is not provided (rare), run `git rev-parse --show-toplevel` as fallback. Never assume CWD is the repo root.
### Decision Recording
After making a decision that affects other team members, write it to:
`.squad/decisions/inbox/{your-name}-{brief-slug}.md`agent-conduct
When your agent uses it
Shared hard rules enforced across all squad agents
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
283 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
Files
SKILL.mdcore1,345 bytes
Read SKILL.md
## Context
Every squad agent must follow these two hard rules. They were previously duplicated in every charter. Now they live here as a shared skill, loaded once.
## Patterns
### Product Isolation Rule (hard rule)
Tests, CI workflows, and product code must NEVER depend on specific agent names from any particular squad. "Our squad" must not impact "the squad." No hardcoded references to cast-specific agent names in test assertions, CI configs, or product logic. Use generic/parameterized values. If a test needs agent names, use obviously-fake test fixtures (e.g., "test-agent-1", "TestBot").
### Peer Quality Check (hard rule)
Before finishing work, verify your changes don't break existing tests. Run the test suite for files you touched. If CI has been failing, check your changes aren't contributing to the problem. When you learn from mistakes, update your history.md.
## Anti-Patterns
- Don't hardcode dev team agent names in product code or testsarchitectural-proposals
When your agent uses it
How to write comprehensive architectural proposals that drive alignment before code is written
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
1,427 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,335 bytes
Read SKILL.md
## Context
Proposals create alignment before code is written. Cheaper to change a doc than refactor code. Use this pattern when:
- Architecture shifts invalidate existing assumptions
- Product direction changes require new foundation
- Multiple waves/milestones will be affected by a decision
- External dependencies (Copilot CLI, SDK APIs) change
## Patterns
### Proposal Structure (docs/proposals/)
**Required sections:**
1. **Problem Statement** — Why current state is broken (specific, measurable evidence)ci-validation-gates
When your agent uses it
Defensive CI/CD patterns: semver validation, token checks, retry logic, and draft detection
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,072 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Known Failure Modes
- Anti-Patterns
Scripts executed separately
../../../scripts/bump-build.mjs
Files
../../../scripts/bump-build.mjsruns4,324 bytesSKILL.mdcore4,562 bytes
Read SKILL.md
## Context
CI workflows must be defensive. These patterns capture lessons from prior release incidents;
they are maintained as version-agnostic gates rather than claims about a particular repository
release.
## Patterns
### Semver Validation Gate
Every publish workflow MUST validate version format before `npm publish`. 4-part versions (e.g., 0.8.21.4) are NOT valid semver — npm mangles them.
```yaml
- name: Validate semver
run: |When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
491 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Checklist — Adding a New CLI Command
- Wiring Patterns by Command Type
- Common Import Pattern
- History
Files
SKILL.mdcore1,844 bytes
Read SKILL.md
# Skill: CLI Command Wiring
**Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.
## Checklist — Adding a New CLI Command
1. **Create command file** in `packages/squad-cli/src/cli/commands/<name>.ts`
- Export a `run<Name>(cwd, options)` async function (or class with static methods for utility modules)
2. **Add routing block** in `packages/squad-cli/src/cli-entry.ts` inside `main()`:
```ts
if (cmd === '<name>') {
const { run<Name> } = await import('./cli/commands/<name>.js');
// parse args, call function
await run<Name>(process.cwd(), options);client-compatibility
When your agent uses it
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,289 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,719 bytes
Read SKILL.md
## Context
Squad runs on multiple Copilot surfaces (CLI, VS Code, JetBrains, GitHub.com). The coordinator must detect its platform and adapt spawning behavior accordingly. Different tools are available on different platforms, requiring conditional logic for agent spawning, SQL usage, and response timing.
## Patterns
### Platform Detection
Before spawning agents, determine the platform by checking available tools:
1. **CLI mode** — `task` tool is available → full spawning control. Use `task` with `agent_type`, `mode`, `model`, `description`, `prompt` parameters. Collect results via `read_agent`.
2. **VS Code mode** — `runSubagent` or `agent` tool is available → conditional behavior. Use `runSubagent` with the task prompt. Drop `agent_type`, `mode`, and `model` parameters. Multiple subagents in one turn run concurrently (equivalent to background mode). Results return automatically — no `read_agent` needed.
coordinator-init-mode
When your agent uses it
The complete two-phase Init Mode protocol the Squad coordinator runs when no team exists yet in the current repo. Phase 1 = propose the team (no files created, wait for user confirm). Phase 2 = create .squad/ scaffolding, casting state, .gitattributes for merge drivers, and the always-on built-ins (Scribe, Ralph, Rai, Fact Checker). Loaded on demand when the coordinator detects no .squad/team.md exists.
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
1,937 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Phase 1: Propose the Team
- Phase 2: Create the Team
Files
SKILL.mdcore7,868 bytes
Read SKILL.md
> **Load this skill when:** you detect that no `.squad/team.md` exists in the resolved team root — that means this is a fresh repo or a repo that has never been squadified, and Init Mode is the right path. The short stub in `squad.agent.md` tells you to load this skill; the full two-phase protocol lives here.
>
> **⚠<U+FE0F> Eager-execution exception:** Init Mode is the ONE exception to the eager-execution / parallel-fan-out doctrine. Phase 1 MUST end with a user confirmation before any file is created. Do not bypass.
## Phase 1: Propose the Team
No team exists yet. **Propose one — but DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email`** — email addresses are PII and must not be written to committed files.
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see the canonical Casting reference at `.squad/templates/casting-reference.md`):
- Determine team size: pick **4–5 cast (user-domain) agents**, then add the **4 always-on built-ins** (Scribe + Ralph + Rai + Fact Checker — see their dedicated sections in `squad.agent.md`). A typical fresh squad has **8–9 total roster entries**, not 4–5.
- Determine assignment shape from the user's project description.
- Derive resonance signals from the session and repo context.coordinator-response-mode
When your agent uses it
Selecting WHO handles work is the Routing table; selecting HOW they handle it (Direct, Lightweight, Standard, Full) is Response Mode. This skill contains the complete decision table, exemplar prompts for each mode, the Lightweight spawn template, and the upgrade rules. Squad coordinator loads this on demand once routing has identified the agent — to pick the right ceremony level for the task.
What it loads into context
Frontmatter
139 tokensAlways available to the agent
SKILL.md
1,172 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Response Mode Selection
- Direct Mode exemplars
- Lightweight Mode exemplars
- Standard Mode exemplars
- Full Mode exemplars
- Mode upgrade rules
- Lightweight Spawn Template
Files
SKILL.mdcore5,018 bytes
Read SKILL.md
> **Load this skill when:** you have routed work to an agent and need to pick the response mode (Direct / Lightweight / Standard / Full). The 1-line stub in `squad.agent.md` is for awareness; this skill is the full decision table + templates.
## Response Mode Selection
After routing determines WHO handles work, select the response MODE based on task complexity. **Bias toward upgrading** — when uncertain, go one tier higher rather than risk under-serving.
| Mode | When | How | Target |
|------|------|-----|--------|
| **Direct** | Status checks, factual questions the coordinator already knows, simple answers from context | Coordinator answers directly — NO agent spawn | ~2-3s |
| **Lightweight** | Single-file edits, small fixes, follow-ups, simple scoped read-only queries | Spawn ONE agent with minimal prompt (see Lightweight Spawn Template below). Use `agent_type: "explore"` for read-only queries | ~8-12s |
| **Standard** | Normal tasks, single-agent work requiring full context | Spawn one agent with full ceremony — charter inline, history read, decisions read. This is the current default | ~25-35s |
| **Full** | Multi-agent work, complex tasks touching 3+ concerns, "Team" requests | Parallel fan-out, full ceremony, Scribe included | ~40-60s |
## Direct Mode exemplarscoordinator-source-of-truth
When your agent uses it
The complete file-by-file source-of-truth hierarchy for Squad: which files are authoritative, which are derived/append-only, who may write each one, who may read each one, and the precedence rules when they conflict. Squad coordinator loads this on demand when it needs to resolve a write conflict, decide where a piece of state belongs, or answer a 'who owns this file' question.
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
1,380 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- State backend note
- File hierarchy
- Rules
Files
SKILL.mdcore5,941 bytesSKILL.mdcore5,941 bytes
Read SKILL.md
> **Load this skill when:** the coordinator (or any agent) needs to resolve a "where does this state belong?" or "who is allowed to write this file?" question — e.g., when about to write `.squad/decisions.md`, when reviewing whether an agent broke the append-only rule, when answering a user question about Squad's file layout, or when adjudicating a conflict between two files. The short summary in `squad.agent.md` is for routing; this skill is the full reference.
## State backend note
Files below marked as **"Derived / append-only"** are **mutable state** — agents access them with runtime state tools (`squad_state_read`, `squad_state_write`, `squad_state_append`, `squad_state_create_if_absent`, `squad_state_delete`, `squad_state_list`). The runtime decides whether the configured backend stores them on disk, git-native state, or an external provider. Files marked as **"Authoritative"** are **static config** and always live on disk regardless of backend.
## File hierarchy
| File | Status | Who May Write | Who May Read |
|------|--------|---------------|--------------|
| `.github/agents/squad.agent.md` | **Authoritative governance.** All roles, handoffs, gates, and enforcement rules. | Repo maintainer (human) | Squad (Coordinator) |
| `.squad/decisions.md` | **Authoritative decision ledger.** Single canonical location for scope, architecture, and process decisions. | Squad (Coordinator) — append only | All agents |
| `.squad/team.md` | **Authoritative roster.** Current team composition. | Squad (Coordinator) | All agents |
| `.squad/routing.md` | **Authoritative routing.** Work assignment rules. | Squad (Coordinator) | Squad (Coordinator) |cross-machine-coordination
When your agent uses it
Design proposal, not implemented: git-based task queuing between squad machines. For competing watchers, use squad watch work claims (ADR-002).
What it loads into context
Frontmatter
58 tokensAlways available to the agent
SKILL.md
3,211 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Usage
- File Formats
- Security Model
- Configuration
- Examples
- Error Handling
- Monitoring & Debugging
- Integration with Ralph Watch
- Migration from Manual Handoff
- Future Enhancements
- Questions?
Files
SKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytesSKILL.mdcore12,300 bytes
Read SKILL.md
# Skill: Cross-Machine Coordination Pattern
**Skill ID:** `cross-machine-coordination`
**Owner:** Ralph (Work Monitor)
**Squad Integration:** All agents
**Status:** Design proposal — NOT implemented
> ⚠<U+FE0F> **Not implemented.** No Squad release reads `.squad/cross-machine/`,
> the `cross_machine` config block, or `squad:machine-{name}` labels. Ralph
> and `squad watch` do not poll task files, run whitelisted commands, or
> write result files. Everything below describes a proposed design, not
> current behavior. Do not tell users it works, and do not create task files
> expecting another machine to pick them up.
>cross-squad
When your agent uses it
Coordinating work across multiple Squad instances — discovery, delegation, and disambiguation when the user says 'squad' (the product) vs casual English 'group of agents'.
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
2,402 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Disambiguation: "squad" vs ad-hoc agents
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytesSKILL.mdcore10,010 bytes
Read SKILL.md
## Context
> **Read this FIRST any time the user says "squad" as a thing to spawn, delegate to, address, or fan out to** — e.g., *"spawn two squads of designers and devs"*, *"ask the other squad"*, *"delegate to a squad"*. In Squad-PRODUCT vocabulary, "squad" is a **peer** (an independent installation with its own `.squad/`, `team.md`, MCP server, and agents) — NOT a generic English synonym for "team" or "group". Do not fan out raw `task` agents inside your own coordinator context when the user means "another squad". Use the discovery and communication patterns below (and the companion `cross-squad-communication` skill for the actual protocols).
When an organization runs multiple Squad instances (e.g., platform-squad, frontend-squad, data-squad), those squads need to discover each other, share context, and hand off work across repository boundaries. This skill teaches agents how to coordinate across squads without creating tight coupling.
> **Companion skill — for protocol details:** `cross-squad-communication/SKILL.md` covers the four communication patterns (synchronous CLI, read-only knowledge query, git-based async, and GitHub-issue-based delegation) once a peer squad is discovered via the registry below. This skill answers "who?" — the companion answers "how?".
Cross-squad orchestration applies when:
- A task requires capabilities owned by another squad
- An architectural decision affects multiple squads
- A feature spans multiple repositories with different squads
- A squad needs to request infrastructure, tooling, or support from another squad
cross-squad-communication
When your agent uses it
Protocol for sending queries, delegating tasks, and sharing context between independent Squad instances across different repositories
What it loads into context
Frontmatter
149 tokensAlways available to the agent
SKILL.md
4,739 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Universal rule: every `copilot` spawn into a peer squad MUST pass `--agent squad`
- Discovery Protocol
- Platform Compatibility Matrix
- Examples
- Anti-Patterns
- Validation Status
Files
SKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytesSKILL.mdcore19,570 bytes
Read SKILL.md
## Context
When multiple repositories each have their own Squad (AI team), they need to exchange information: knowledge queries, PR reviews, task delegation, and dependency analysis. Each squad has its own agents, MCP tools, and issue tracker — there is no shared runtime.
> **Companion skill — read first:** `cross-squad/SKILL.md` covers **discovery** of peer squads via `squad registry add/list/remove`. This skill picks up after a peer is known and covers the **communication protocols** themselves — the four numbered patterns below: Pattern 0 (synchronous CLI), Pattern 1 (read-only knowledge query), Pattern 2 (git-based async), and Pattern 3 (GitHub-issue-based delegation). A separate non-numbered appendix (Cross-Repo Dependency Scan) is provided as a related analysis tool, not a communication pattern. The two skills are designed to be used together.
**When this skill applies:**
- A squad agent needs information from another squad-enabled repo
- A task needs to be delegated to another squad
- Cross-repo dependency analysis is needed
- PR review requests span repo boundaries
**Key constraint:** Each squad has its own runtime, MCP tools, and issue tracker. Cross-squad communication can be **synchronous** (via CLI session targeting the other repo) or **asynchronous** (file-based or issue-based). The coordinator decides which approach fits.
distributed-mesh
When your agent uses it
How to coordinate with squads on different machines using git as transport
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
3,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Patterns
- Examples
- AGENT WORKFLOW (Deterministic Setup)
- Anti-Patterns
Scripts executed separately
sync-mesh.ps1sync-mesh.sh
Files
SKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesSKILL.mdcore12,427 bytesmesh.json.exampleasset762 bytessync-mesh.ps1runs3,861 bytessync-mesh.shruns3,395 bytes
Read SKILL.md
## SCOPE
**✅ THIS SKILL PRODUCES (exactly these, nothing more):**
1. **`mesh.json`** — Generated from user answers about zones and squads (which squads participate, what zone each is in, paths/URLs for each), using `mesh.json.example` in this skill's directory as the schema template
2. **`sync-mesh.sh` and `sync-mesh.ps1`** — Copied from this skill's directory into the project root (these are bundled resources, NOT generated code)
3. **Zone 2 state repo initialization** (if applicable) — If the user specified a Zone 2 shared state repo, run `sync-mesh.sh --init` to scaffold the state repo structure
4. **A decision entry** in `.squad/decisions/inbox/` documenting the mesh configuration for team awareness
**❌ THIS SKILL DOES NOT PRODUCE:**
- **No application code** — No validators, libraries, or modules of any kind
- **No test files** — No test suites, test cases, or test scaffolding
- **No GENERATING sync scripts** — They are bundled with this skill as pre-built resources. COPY them, don't generate them.docs-standards
When your agent uses it
Microsoft Style Guide + Squad-specific documentation patterns
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
524 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore2,270 bytesSKILL.mdcore2,270 bytes
Read SKILL.md
## Context
Squad documentation follows the Microsoft Style Guide with Squad-specific conventions. Consistency across docs builds trust and improves discoverability.
## Patterns
### Microsoft Style Guide Rules
- **Sentence-case headings:** "Getting started" not "Getting Started"
- **Active voice:** "Run the command" not "The command should be run"
- **Second person:** "You can configure..." not "Users can configure..."
- **Present tense:** "The system routes..." not "The system will route..."
- **No ampersands in prose:** "and" not "&" (except in code, brand names, or UI elements)
### Squad Formatting Patternse2e-template-testing
When your agent uses it
End-to-end validation of coordinator and agent template changes
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
6,164 tokensWhen the skill is activated
Reference documents
5,121 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- When To Use
- Prerequisites
- Workflow
- Progress Reporting
- Test Matrix Template
- Tips
- Fast-Fail Rules
- PII Protection — Mandatory
- Anti-Patterns
- Sandbox / Permission Notes
- Confidence
Files
../../../CONTRIBUTING.mddoc21,149 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytesSKILL.mdcore23,737 bytes
Read SKILL.md
## Context
Squad's coordinator prompt (`squad.agent.md`) and agent charters (e.g.
`scribe-charter.md`) are shipped as templates in `.squad-templates/`. Changes to
these files affect how every squad session behaves — but unit tests can't catch
prompt-level regressions because the prompts are interpreted by an LLM at
runtime.
This skill describes how to validate template changes end-to-end by running real
squad sessions against a locally-built CLI that includes your modified templates.
## When To Use
- You changed `.squad-templates/squad.agent.md` (coordinator prompt)economy-mode
When your agent uses it
Shifts Layer 3 model selection to cost-optimized alternatives when economy mode is active.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,396 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- Activation Methods
- Economy Model Selection Table
- AGENT WORKFLOW
- Config Schema
- Anti-Patterns
Files
SKILL.mdcore5,547 bytes
Read SKILL.md
## SCOPE
✅ THIS SKILL PRODUCES:
- A modified Layer 3 model selection table applied when economy mode is active
- `economyMode: true` written to `.squad/config.json` when activated persistently
- Spawn acknowledgments with `💰` indicator when economy mode is active
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Cost reports or billing artifacts
- Changes to Layer 0, Layer 1, or Layer 2 resolution (user intent always wins)
## Context
error-recovery
When your agent uses it
Standard recovery patterns for all squad agents. When something fails, adapt — don't just report the failure.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
785 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Retry with Backoff
- 2. Fallback Alternatives
- 3. Diagnose-and-Fix
- 4. Escalate with Context
- 5. Graceful Degradation
- Applying These Patterns
Files
SKILL.mdcore3,466 bytesSKILL.mdcore3,466 bytes
Read SKILL.md
# Error Recovery Patterns
Standard recovery patterns for all squad agents. When something fails, **adapt** — don't just report the failure.
---
## 1. Retry with Backoff
**When:** Transient failures — API timeouts, rate limits, network errors, temporary service unavailability.
**Pattern:**
1. Wait briefly, then retry (start at 2s, double each attempt)
2. Maximum 3 retries before escalating
3. Log each attempt with the error receivedexternal-comms
When your agent uses it
Workflow for scanning, drafting, and presenting community responses with a human review gate
What it loads into context
Frontmatter
228 tokensAlways available to the agent
SKILL.md
2,993 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore12,847 bytesSKILL.mdcore12,847 bytes
Read SKILL.md
## Context
Phase 1 is **draft-only mode**.
- The communications owner scans issues and discussions, drafts responses with the humanizer skill, and presents a review table for human approval.
- **Human review gate is mandatory** — the communications owner never posts autonomously.
- Every action is logged to `.squad/comms/audit/`.
- This workflow is triggered manually only ("check community") — no automated activation in Phase 1.
## Patterns
### 1. Scan
Find unanswered community items with GitHub MCP tools first, or `gh issue list` / `gh api` as fallback for issues and discussions.fact-checking
When your agent uses it
Review and validate claims using counter-hypothesis testing. Use when verifying technical content, checking references, validating API endpoints, or performing quality assurance on deliverables.
What it loads into context
Frontmatter
97 tokensAlways available to the agent
SKILL.md
469 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore2,046 bytes
Read SKILL.md
# Skill: Fact Checking
## Context
Codifies the challenger agent review output format and methodology so any agent performing fact-checking or review produces consistent, structured output.
## Pattern
### Review Methodology
For every claim or deliverable under review:
1. Ask: "What evidence supports this? What would disprove it?"
2. Generate counter-hypotheses and test them against available data
3. Verify URLs, package names, API endpoints, and external references actually exist
4. Flag confidence levels: ✅ Verified, ⚠<U+FE0F> Unverified, ❌ Contradictedgh-auth-isolation
When your agent uses it
Safely manage multiple GitHub identities (EMU + personal) in agent workflows
What it loads into context
Frontmatter
100 tokensAlways available to the agent
SKILL.md
1,537 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytesSKILL.mdcore6,441 bytes
Read SKILL.md
## Context
Many developers use GitHub through an Enterprise Managed User (EMU) account at work while maintaining a personal GitHub account for open-source contributions. AI agents spawned by Squad inherit the shell's default `gh` authentication — which is usually the EMU account. This causes failures when agents try to push to personal repos, create PRs on forks, or interact with resources outside the enterprise org.
This skill teaches agents how to detect the active identity, switch contexts safely, and avoid mixing credentials across operations.
## Patterns
### Detect Current Identity
Before any GitHub operation, check which account is active:
```bash
gh auth statusgh-aw-enlistment
When your agent uses it
Enlist a repository into Squad by installing the supported GitHub Agentic Workflows (gh-aw) bootstrap — with strict compilation, an explicit safe-update allowlist, and a human-reviewed bootstrap PR. Never auto-merges.
What it loads into context
Frontmatter
308 tokensAlways available to the agent
SKILL.md
7,239 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytes
Read SKILL.md
## Context
Use this skill when a user wants to **enlist a repository into Squad** through
**GitHub Agentic Workflows (`gh aw`)** — e.g. "set up Squad agentic workflows",
"enlist this repo in Squad", or "install Squad gh-aw workflows". The end state is
the `/squad` slash command being live on the repo, delivered as a **human-reviewed
bootstrap pull request** — never an auto-merge.
This skill **operationalizes** the supported bootstrap path. It does not summarize
it: each step below is a gate with explicit success evidence and **STOP conditions**.
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this
skill ever disagree, the guide wins — re-read it before proceeding.
git-workflow
When your agent uses it
Squad branching model: dev-first workflow with insiders preview channel
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,581 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Branch Naming Convention
- Workflow for Issue Work
- Parallel Multi-Issue Work (Worktrees)
- Multi-Repo Downstream Scenarios
- Anti-Patterns
- Promotion Pipeline
Files
SKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytesSKILL.mdcore6,259 bytes
Read SKILL.md
## Context
Squad uses a three-branch model. **All feature work starts from `dev`, not `main`.**
| Branch | Purpose | Publishes |
|--------|---------|-----------|
| `main` | Released, tagged, in-npm code only | `npm publish` on tag |
| `dev` | Integration branch — all feature work lands here | `npm publish --tag preview` on merge |
| `insiders` | Early-access channel — synced from dev | `npm publish --tag insiders` on sync |
## Branch Naming Convention
Issue branches MUST use: `squad/{issue-number}-{kebab-case-slug}`
github-multi-account
When your agent uses it
Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
What it loads into context
Frontmatter
80 tokensAlways available to the agent
SKILL.md
1,038 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Activate
- What to Do (as the AI agent)
- After Setup — Usage Rules
- Repo-Specific Account Binding
- For Squad Agents
Files
SKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytesSKILL.mdcore3,609 bytes
Read SKILL.md
# GitHub Multi-Account — AI-Driven Setup
## When to Activate
When the user has multiple GitHub accounts (check with `gh auth status`). If you see 2+ accounts listed, this skill applies.
## What to Do (as the AI agent)
### Step 1: Detect accounts
Run: `gh auth status`
Look for multiple accounts. Note which usernames are listed.
### Step 2: Ask the user
Ask: "I see you have multiple GitHub accounts: {list them}. Which one is your personal account and which is your work/EMU account?"
history-hygiene
When your agent uses it
Record reconciled final outcomes to history.md, not intermediate requests or obsolete claims
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
545 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Scope
- Patterns
- Examples
- Stop conditions
- Anti-Patterns
Files
SKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytesSKILL.mdcore2,723 bytes
Read SKILL.md
## Context
History files are cold-start context, not the source of truth for live behavior. Stale or incorrect
entries poison decision-making downstream. A stale-history incident proved that an old history claim
can cause future spawns to repeat a reversed decision. Reconcile an operational claim with current
implementation, workflows, tests, documentation, skills, and applicable decisions before acting on it.
## Scope
- Activate when recording, correcting, or promoting finalized operational history.
- Allowed targets: the explicitly selected history file and a focused reusable skill when the rule
applies beyond one role.
- Exclusions: intermediate requests, stale assignments, roster metadata, and unrelated cleanup.
humanizer
When your agent uses it
Tone enforcement patterns for external-facing community responses
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
1,004 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,330 bytes
Read SKILL.md
## Context
Use this skill whenever the communications owner drafts external-facing responses for issues or discussions.
- Tone must be warm, helpful, and human-sounding — never robotic or corporate.
- The repository's communication policy applies everywhere: **Humanized tone is mandatory**.
- This applies to **all external-facing content** drafted by the communications owner in Phase 1 issues/discussions workflows.
## Patterns
1. **Warm opening** — Start with acknowledgment ("Thanks for reporting this", "Great question!")
2. **Active voice** — "We're looking into this" not "This is being investigated"
3. **Second person** — Address the person directly ("you" not "the user")
4. **Conversational connectors** — "That said...", "Here's what we found...", "Quick note:"init-mode
When your agent uses it
Team initialization flow (Phase 1 proposal + Phase 2 creation)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
1,749 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,100 bytesSKILL.mdcore7,100 bytes
Read SKILL.md
## Context
Init Mode activates when `.squad/team.md` does not exist, or exists but has zero roster entries under `## Members`. The coordinator proposes a team (Phase 1), waits for user confirmation, then creates the team structure (Phase 2).
## Patterns
### Phase 1: Propose the Team
No team exists yet. Propose one — but **DO NOT create any files until the user confirms.**
1. **Identify the user.** Run `git config user.name` to learn who you're working with. Use their name in conversation (e.g., *"Hey {user}, what are you building?"*). Store their name (NOT email) in `team.md` under Project Context. **Never read or store `git config user.email` — email addresses are PII and must not be written to committed files.**
2. Ask: *"What are you building? (language, stack, what it does)"*
3. **Cast the team.** Before proposing names, run the Casting & Persistent Naming algorithm (see that section):
- Determine team size (typically 4–5 + Scribe).iterative-retrieval
When your agent uses it
Max-3-cycle protocol for agent sub-tasks with WHY context and coordinator validation. Use when spawning sub-agents to complete scoped work.
What it loads into context
Frontmatter
55 tokensAlways available to the agent
SKILL.md
1,385 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Spawn Prompt Template
- 3-Cycle Protocol
- Coordinator Validation Checklist
- When to Escalate vs Retry
- Issue Dedup Check (Mandatory)
- Mandatory Output Requirement (Research-Then-Execute)
- Anti-Patterns
- Examples
Files
SKILL.mdcore6,341 bytesSKILL.mdcore6,341 bytes
Read SKILL.md
# Iterative Retrieval Skill
Squad agents frequently spawn sub-agents to complete scoped work. Without structure, these
handoffs become vague, cycles multiply, and outputs land without being checked. The
**Iterative Retrieval Pattern** caps cycles at 3, mandates WHY context in every spawn, and
requires the coordinator to validate agent output before closing an issue.
---
## Spawn Prompt Template
Every agent spawn must include the following four sections. Copy and fill in the template:
```model-selection
When your agent uses it
Determines which LLM model to use for each agent spawn
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- SCOPE
- Context
- 5-Layer Model Resolution Hierarchy
- AGENT WORKFLOW
- Config Schema
- Fallback Chains
Files
SKILL.mdcore8,082 bytes
Read SKILL.md
# Model Selection
> Determines which LLM model to use for each agent spawn.
## SCOPE
✅ THIS SKILL PRODUCES:
- A resolved `model` parameter for every `task` tool call
- Persistent model preferences in `.squad/config.json`
- Spawn acknowledgments that include the resolved model
❌ THIS SKILL DOES NOT PRODUCE:
- Code, tests, or documentation
- Model performance benchmarksnap
When your agent uses it
Context hygiene — compress, prune, archive .squad/ state
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
208 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What It Does
- When To Use
- Invocation
- Confidence
Files
SKILL.mdcore792 bytes
Read SKILL.md
# Skill: nap
> Context hygiene — compress, prune, archive .squad/ state
## What It Does
Reclaims context window budget by compressing agent histories, pruning old logs,
archiving stale decisions, and cleaning orphaned inbox files.
## When To Use
- Before heavy fan-out work (many agents will spawn)
- When history.md files exceed 15KB
- When .squad/ total size exceeds 1MBnotification-routing
When your agent uses it
Route agent notifications to specific channels by type — prevent alert fatigue from single-channel flooding
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
730 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Anti-Patterns
- Distributed Systems Pattern
Files
SKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytesSKILL.mdcore3,084 bytes
Read SKILL.md
## Context
When a Squad grows beyond a few agents, notifications flood a single channel — failure alerts drown in daily
briefings, tech news buries security findings, and everything gets ignored. This is the pub-sub problem:
a single message queue for everything is a recipe for missed alerts.
The fix is **topic-based routing**: agents tag notifications with a channel type, and a routing function
sends them to the appropriate destination.
**Trigger symptoms:**
- Important alerts missed because they're buried in routine notifications
- Team members turning off notifications entirely (signal overwhelm)
- Onboarding friction: "where do I look for X?"
pr-review-response
When your agent uses it
Teaches agents to reply to PR review comment threads after fixing issues, making resolutions traceable
What it loads into context
Frontmatter
154 tokensAlways available to the agent
SKILL.md
2,574 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- SCOPE
- Patterns
- AGENT WORKFLOW (Summary)
- Examples
- Anti-Patterns
Files
SKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytesSKILL.mdcore10,724 bytes
Read SKILL.md
## Context
When an agent fixes code in response to PR review comments (from Copilot, a human reviewer, or any GitHub reviewer), the fix alone is not enough. The reviewer needs to see — on the PR thread itself — which comments were addressed and how. Without replies, comments stay visually unresolved, reviewers must re-read the entire diff to verify fixes, and there's no traceable link between feedback and resolution.
Use this skill whenever:
- You are fixing code based on PR review feedback
- You are addressing Copilot review suggestions
- You are responding to reviewer-requested changes on a PR
- A squad member hands you review comments to resolve
## SCOPE
✅ THIS SKILL PRODUCES:
- Reply comments on each review thread explaining the fixpr-screenshots
When your agent uses it
Capture Playwright screenshots and embed them in GitHub PR descriptions
What it loads into context
Frontmatter
62 tokensAlways available to the agent
SKILL.md
1,058 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../scripts/capture-pr-screenshots.mjs
Files
../../../scripts/capture-pr-screenshots.mjsruns1,855 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytesSKILL.mdcore4,455 bytes
Read SKILL.md
## Context
When a PR includes visual changes (docs sites, UI components, generated pages), reviewers
need to see what the PR delivers without checking out the branch. Screenshots belong in
the **PR description body**, not as committed files and not as text descriptions.
Use this skill whenever:
- A PR touches docs site pages (Astro, Starlight, etc.)
- A PR adds or changes UI components
- A PR generates visual artifacts (TypeDoc, Storybook, diagrams)
- Playwright tests already capture screenshots as part of testing
## Patterns
project-conventions
When your agent uses it
Core conventions and patterns for this codebase
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,555 bytes
Read SKILL.md
## Context
> **This is a starter template.** Replace the placeholder patterns below with your actual project conventions. Skills train agents on codebase-specific practices — accurate documentation here improves agent output quality.
## Patterns
### [Pattern Name]
Describe a key convention or practice used in this codebase. Be specific about what to do and why.
### Error Handling
<!-- Example: How does your project handle errors? -->
<!-- - Use try/catch with specific error types? -->ralph-two-pass-scan
When your agent uses it
Cuts GitHub API calls by separating lightweight list scanning from full hydration
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
345 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Pattern
Files
SKILL.mdcore1,336 bytes
Read SKILL.md
# Skill: Ralph — Two-Pass Issue Scanning
**Confidence:** high
**Domain:** work-monitoring
**Last validated:** 2026-03-24
## Context
Cuts GitHub API calls from N+1 to ~7 per round (~72% reduction) by separating list scanning from full hydration.
Addresses the scanning inefficiency described in issue #596.
## Pattern
### Pass 1 — Lightweight Scan
```reflect
When your agent uses it
Learning capture system that extracts HIGH/MED/LOW confidence patterns from conversations to prevent repeating mistakes. Use after user corrections ("no", "wrong"), praise ("perfect", "exactly"), or when discovering edge cases. Complements .squad/agents/{agent}/history.md and .squad/decisions.md.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Integration with Squad Architecture
- Triggers
- Process
- Usage Examples
- When to Use
- See Also
Files
SKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytesSKILL.mdcore8,794 bytes
Read SKILL.md
# Reflect Skill
**Critical learning capture system** for Squad. Prevents repeating mistakes and preserves successful patterns across sessions.
Analyze conversations and propose improvements to squad knowledge based on what worked, what didn't, and edge cases discovered. **Every correction is a learning opportunity.**
---
## Integration with Squad Architecture
**Reflect complements existing Squad knowledge systems:**
1. **Governed runtime memory** — Durable observations and decisions persisted through configured state tools
2. **Tracked governing source** — Coordinator prompts, skills, and other behavior-defining files changed through focused pull requestsrelease-process
When your agent uses it
Prepare, publish, recover, and verify Squad insider, preview, and stable releases
What it loads into context
Frontmatter
41 tokensAlways available to the agent
SKILL.md
2,578 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Non-negotiable rules
- Required credentials
- Human release-trigger boundary
- Prepare a release
- Publish a preview release
- Publish an insider release
- Publish a stable release
- Verify publication
- Recovery
- After a stable release
Files
SKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytesSKILL.mdcore10,306 bytes
Read SKILL.md
# Release Process
This is the canonical Squad release runbook. Squad has three release channels
and two long-lived branches:
| Source | Version | GitHub | npm | Standalone | Homebrew/WinGet |
|--------|---------|--------|-----|------------|-----------------|
| Insider dispatch from `dev` | Generated `X.Y.Z-insider.N` | Prerelease | `insider` | Yes | Yes |
| Release dispatch from `dev` | `X.Y.Z-preview.N` | Prerelease | `preview` | Yes | Yes |
| Promotion PR merged to `main` (or legacy workflow push) | `X.Y.Z` | Stable/latest | `latest` | Yes | Yes |
The repository does not use a staging `preview` branch. Preview is a release
channel, not a branch.
reskill
When your agent uses it
Extract durable operating knowledge before slimming or replacing team charters
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,223 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Process
- Patterns
- Anti-Patterns
Files
SKILL.mdcore5,760 bytesSKILL.mdcore5,760 bytes
Read SKILL.md
## Context
When the coordinator hears "team, reskill" (or similar: "optimize context", "slim down charters"), trigger a team-wide optimization pass. The goal: reduce per-agent context consumption by extracting shared patterns from charters and histories into reusable skills.
This is a periodic maintenance activity. Run whenever charter/history bloat is suspected or a
team will be re-cast. Charters and histories are evidence, not the authority for current product
behavior: reconcile candidates against current source, workflows, tests, documentation, skills,
and decisions before preserving them.
## Process
### Step 1: Audit
Read all active agent charters and histories, then inventory existing skills. Measure byte sizes.
Identify:When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
1,352 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Purpose
- Core Function: Test-RetroOverdue
- Coordinator Integration
- Action Item Enforcement
- Cadence Enforcement
- Skill Metadata
Files
SKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytesSKILL.mdcore4,938 bytes
Read SKILL.md
# Skill: Retro Enforcement
## Purpose
Ensure retrospectives happen on schedule and that their action items are tracked in GitHub Issues — not markdown checklists.
This skill addresses a specific, measured failure mode: **0% completion rate on markdown retro action items across 6 consecutive retrospectives**. GitHub Issues have an 85%+ completion rate in the same squad. The format was the problem, not the people.
## Core Function: Test-RetroOverdue
```powershell
function Test-RetroOverdue {
param(
[string]$LogDir = ".squad/log",
[int]$WindowDays = 7,reviewer-protocol
When your agent uses it
Reviewer rejection workflow and strict lockout semantics
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
1,009 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore4,658 bytes
Read SKILL.md
## Context
When a team member has a **Reviewer** role (e.g., Tester, Code Reviewer, Lead), they may approve or reject work from other agents. On rejection, the coordinator enforces strict lockout rules to ensure the original author does NOT self-revise. This prevents defensive feedback loops and ensures independent review.
## Patterns
### Reviewer Rejection Protocol
Quality and test reviewers may block a PR or merge on quality, coverage, or regression grounds. Treat that block as a rejection that enters the protocol below.
When a team member has a **Reviewer** role:
- Reviewers may **approve** or **reject** work from other agents.
- On **rejection**, the Reviewer may choose ONE of:secret-handling
When your agent uses it
Never read .env files or write secrets to .squad/ committed files
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
1,994 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore7,524 bytesSKILL.mdcore7,524 bytes
Read SKILL.md
## Context
Spawned agents have read access to the entire repository, including `.env` files containing live credentials. If an agent reads secrets and writes them to `.squad/` files (decisions, logs, history), Scribe auto-commits them to git, exposing them in remote history. This skill codifies absolute prohibitions and safe alternatives.
## Patterns
### Prohibited File Reads
**NEVER read these files:**
- `.env` (production secrets)
- `.env.local` (local dev secrets)
- `.env.production` (production environment)
- `.env.development` (development environment)
- `.env.staging` (staging environment)session-recovery
When your agent uses it
Find and resume interrupted Copilot CLI sessions using session_store queries
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,215 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytesSKILL.mdcore5,174 bytes
Read SKILL.md
## Context
Squad agents run in Copilot CLI sessions that can be interrupted — terminal crashes, network drops, machine restarts, or accidental window closes. When this happens, in-progress work may be left in a partially-completed state: branches with uncommitted changes, issues marked in-progress with no active agent, or checkpoints that were never finalized.
Copilot CLI stores session history in a SQLite database called `session_store` (read-only, accessed via the `sql` tool with `database: "session_store"`). This skill teaches agents how to query that store to detect interrupted sessions and resume work.
## Patterns
### 1. Find Recent Sessions
Query the `sessions` table filtered by time window. Include the last checkpoint to understand where the session stopped:
```sql
SELECTsquad
When your agent uses it
Squad's command catalog and interactive menu. Invoke via /squad (slash command) or natural language ("squad commands", "what can squad do", "show me squad options"). Presents categorized operations (Install & Upgrade, Team Management, Issues & PRs, Plugins & Skills, Model & Cost, Sessions & State) as an interactive picker. Routes to the right squad CLI command or the Squad coordinator agent.
What it loads into context
Frontmatter
103 tokensAlways available to the agent
SKILL.md
2,680 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,874 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill (via `/squad` slash command, "squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-commands
When your agent uses it
Categorized catalog of common Squad operations. Coordinator reads this file and presents it as an interactive menu when the user asks for available commands or help.
What it loads into context
Frontmatter
82 tokensAlways available to the agent
SKILL.md
2,652 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Menu Presentation Rules
- Install & Upgrade
- Team Management
- Issues & PRs
- Plugins & Skills
- Model & Cost
- Sessions & State
Files
SKILL.mdcore10,748 bytes
Read SKILL.md
## Menu Presentation Rules
When the user triggers this skill ("squad commands", "help", "what can squad do", etc.):
1. **Category-level menu first.** Present category names as an `ask_user` choice list:
```
📋 Squad Commands — pick a category:
1. Install & Upgrade
2. Team Management
3. Issues & PRs
4. Plugins & Skills
5. Model & Cost
6. Sessions & State
```squad-conventions
When your agent uses it
Core conventions and patterns used in the Squad codebase
What it loads into context
Frontmatter
39 tokensAlways available to the agent
SKILL.md
824 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore3,440 bytesSKILL.mdcore3,440 bytes
Read SKILL.md
## Context
These conventions apply to all work on the Squad CLI tool (`create-squad`). Squad is a zero-dependency Node.js package that adds AI agent teams to any project. Understanding these patterns is essential before modifying any Squad source code.
## Patterns
### Zero Dependencies
Squad has zero runtime dependencies. Everything uses Node.js built-ins (`fs`, `path`, `os`, `child_process`). Do not add packages to `dependencies` in `package.json`. This is a hard constraint, not a preference.
### Node.js Built-in Test Runner
Tests use `node:test` and `node:assert/strict` — no test frameworks. Run with `npm test`. Test files live in `test/`. The test command is `node --test test/`.
### Error Handling — `fatal()` Pattern
All user-facing errors use the `fatal(msg)` function which prints a red `✗` prefix and exits with code 1. Never throw unhandled exceptions or print raw stack traces. The global `uncaughtException` handler calls `fatal()` as a safety net.
squad-help
When your agent uses it
How to actually use Squad — Squad is a custom Copilot agent (invoked via the task tool with agent_type='Squad'), not a skill. This file explains the right invocation paths for setting up a team, listing squad commands, and initializing Squad in a new project.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,043 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Squad is a custom agent, not a skill
- How to actually use Squad
- What NOT to do
- How this skill was discovered
- See also
Files
SKILL.mdcore4,198 bytesSKILL.mdcore4,198 bytes
Read SKILL.md
# Skill: squad-help
> **Quick reference.** If you're reading this because a user said "use squad" or "squad" or "set up a squad", you're in the right place — read on for the correct invocation paths.
---
## Squad is a custom agent, not a skill
The Squad framework registers a **custom Copilot CLI agent** at `.github/agents/squad.agent.md`. The agent is named **`Squad`** and its description is *"Your AI team. Describe what you're building, get a team of specialists that live in your repo."*
Copilot CLI agents and skills are different things:
| Thing | How to invoke | Example |
|---|---|---|squad-version-check
When your agent uses it
Internals of how @bradygaster/squad-cli stamps its version, how `squad upgrade` works (what it preserves vs overwrites), and how to probe the npm registry for the latest version from a coordinator prompt.
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,702 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What This Skill Covers
- Package & Registry Facts
- Version Stamping Mechanism
- `squad upgrade` Behavior
- `.squad/config.json` — What It Holds
- Version-Check Probe (npm Registry)
- Upstream OS-Specific Cache
- Repo-Local Cache Convention: `.squad/.cache/version-check.json`
- Key File Paths (installed CLI)
Files
SKILL.mdcore6,296 bytes
Read SKILL.md
# SKILL: Squad CLI Internals — Version Stamping & Upgrade Mechanics
**Confidence:** medium
**Discovered by:** repository investigation
**Date:** 2026-05-26
**Validated in:** Issue #1173 recon (bradygaster/squad)
---
## What This Skill Covers
Reusable knowledge about how `@bradygaster/squad-cli` stamps its version into `squad.agent.md`, how `squad upgrade` works, what it preserves vs. overwrites, and how to probe the npm registry for the latest version from a coordinator prompt.
---test-discipline
When your agent uses it
Design assertions that prove the actual contract and fail on realistic regressions
What it loads into context
Frontmatter
40 tokensAlways available to the agent
SKILL.md
346 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytesSKILL.mdcore1,787 bytes
Read SKILL.md
## Context
Use this skill whenever a change affects an API, public interface, workflow contract, generated
artifact, or user-observable behavior. A passing status-only test is not evidence that the
intended artifact or contract is correct.
## Patterns
- Update tests in the same change as an API, signature, export, or public-interface change.
- Make assertions name and inspect the offending artifact, output, or configuration field—not just
an exit code or broad success status.
- Mutation-test critical gates by changing the real source or emitted artifact and proving the
assertion turns red.
- Exercise every caller path for a changed parser, normalizer, or helper; isolated unit coveragetiered-memory
When your agent uses it
Three-tier agent memory model (hot/cold/wiki) for context reduction per spawn
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
1,775 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Memory Tiers
- When to Load Each Tier
- Spawn Template Pattern
- Integration with Scribe Agent (design — not yet implemented)
- Implementation Checklist (tracked in #1264)
- References
- Spawn Template
- Task
- WHY
- Success Criteria
- Memory Context
- Escalation
- Notes
Files
SKILL.mdcore7,506 bytes
Read SKILL.md
# Skill: Tiered Agent Memory
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, and spawn-template tier-aware reads are tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264). Until those land, agents continue to load full `history.md` + `decisions.md` on every spawn.
## Overview
Squad agents today load their full context history on every spawn, which grows unboundedly across sessions. The Tiered Agent Memory model proposes a three-tier separation so agents only load the bytes that are actually relevant to the current task, with older context kept available on demand.
---
## Memory Tiers
### 🔥 Hot Tier — Current Session Context
- **Size target:** keep small (~2–4KB typical)versioning-policy
When your agent uses it
SemVer rules for Squad's stable, preview, insider, and local package versions
What it loads into context
Frontmatter
60 tokensAlways available to the agent
SKILL.md
1,153 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- 1. Supported versions
- 2. Package versions stay in lockstep
- 3. Prerelease workspace dependency rule
- 4. Local build versions are ephemeral
- 5. Release lifecycle
- 6. Ownership
- 7. CI enforcement
- Quick reference
Scripts executed separately
../../../scripts/bump-build.mjs
Files
../../../scripts/bump-build.mjsruns4,324 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytesSKILL.mdcore4,783 bytes
Read SKILL.md
## Context
Squad publishes `@bradygaster/squad-sdk` and
`@bradygaster/squad-cli` from one npm workspace. Their versions and the root
version move together. A mismatched prerelease dependency can make npm silently
resolve an older registry SDK instead of the local workspace.
## 1. Supported versions
| Use | Version | Committed branch |
|---|---|---|
| Stable release | `MAJOR.MINOR.PATCH` | `main`, and briefly `dev` during promotion |
| Preview release | `MAJOR.MINOR.PATCH-preview.N` | `dev` |
| Insider snapshot | `MAJOR.MINOR.PATCH-insider.N` | Generated by the insider workflow |windows-compatibility
When your agent uses it
Cross-platform path handling and command patterns
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
1,572 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Scripts executed separately
../../../scripts/fix-crlf-worktree.mjs
Files
../../../scripts/fix-crlf-worktree.mjsruns5,444 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytesSKILL.mdcore6,389 bytes
Read SKILL.md
## Context
Squad runs on Windows, macOS, and Linux. Several bugs have been traced to platform-specific assumptions: ISO timestamps with colons (illegal on Windows), `git -C` with Windows paths (unreliable), forward-slash paths in Node.js on Windows.
## Patterns
### Filenames & Timestamps
- **Never use colons in filenames:** ISO 8601 format `2026-03-15T05:30:00Z` is illegal on Windows
- **Use `safeTimestamp()` utility:** Replaces colons with hyphens → `2026-03-15T05-30-00Z`
- **Centralize formatting:** Don't inline `.toISOString().replace(/:/g, '-')` — use the utility
### Git Commands
- **Never use `git -C {path}`:** Unreliable with Windows paths (backslashes, spaces, drive letters)
- **Always `cd` first:** Change directory, then run git commandsWhen your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
0 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
SKILL.mdcore0 bytes
My Skill
When your agent uses it
What it loads into context
Frontmatter
21 tokensAlways available to the agent
SKILL.md
30 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
SKILL.mdcore98 bytes
gh-aw-enlistment
When your agent uses it
Enlist a repository into Squad by installing the supported GitHub Agentic Workflows (gh-aw) bootstrap — with strict compilation, an explicit safe-update allowlist, and a human-reviewed bootstrap PR. Never auto-merges.
What it loads into context
Frontmatter
308 tokensAlways available to the agent
SKILL.md
7,239 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Context
- Patterns
- Examples
- Anti-Patterns
Files
SKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytesSKILL.mdcore30,292 bytes
Read SKILL.md
## Context
Use this skill when a user wants to **enlist a repository into Squad** through
**GitHub Agentic Workflows (`gh aw`)** — e.g. "set up Squad agentic workflows",
"enlist this repo in Squad", or "install Squad gh-aw workflows". The end state is
the `/squad` slash command being live on the repo, delivered as a **human-reviewed
bootstrap pull request** — never an auto-merge.
This skill **operationalizes** the supported bootstrap path. It does not summarize
it: each step below is a gate with explicit success evidence and **STOP conditions**.
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this
skill ever disagree, the guide wins — re-read it before proceeding.
Skills included
- action-skill-contracts
- agent-collaboration
- agent-conduct
- architectural-proposals
- architectural-review
- archival-integrity
- ci-validation-gates
- cli-wiring
- client-compatibility
- content-scope
- distributed-mesh
- documentation-maintenance
- git-workflow
- github-multi-account
- history-hygiene
- init-mode
- model-selection
- pr-lifecycle
- protected-files
- release-process
- reskill
- reviewer-protocol
- secret-handling
- security-review
- squad-conventions
- test-discipline
- windows-compatibility
- {skill-name}
- agent-collaboration
- agent-conduct
- architectural-proposals
- ci-validation-gates
- client-compatibility
- coordinator-init-mode
- coordinator-response-mode
- coordinator-source-of-truth
- cross-machine-coordination
- cross-squad
- cross-squad-communication
- distributed-mesh
- docs-standards
- e2e-template-testing
- economy-mode
- error-recovery
- external-comms
- fact-checking
- gh-auth-isolation
- gh-aw-enlistment
- git-workflow
- github-multi-account
- history-hygiene
- humanizer
- init-mode
- iterative-retrieval
- model-selection
- nap
- notification-routing
- pr-review-response
- pr-screenshots
- project-conventions
- ralph-two-pass-scan
- reflect
- release-process
- reskill
- reviewer-protocol
- secret-handling
- session-recovery
- squad
- squad-commands
- squad-conventions
- squad-help
- squad-version-check
- test-discipline
- tiered-memory
- versioning-policy
- windows-compatibility
- model-selection
- team-memory-maintenance
- work-monitoring
- {skill-name}
- {skill-name}
- agent-collaboration
- agent-conduct
- architectural-proposals
- ci-validation-gates
- client-compatibility
- coordinator-init-mode
- coordinator-response-mode
- coordinator-source-of-truth
- cross-machine-coordination
- cross-squad
- cross-squad-communication
- distributed-mesh
- docs-standards
- e2e-template-testing
- economy-mode
- error-recovery
- external-comms
- fact-checking
- gh-auth-isolation
- gh-aw-enlistment
- git-workflow
- github-multi-account
- history-hygiene
- humanizer
- init-mode
- iterative-retrieval
- model-selection
- nap
- notification-routing
- pr-review-response
- pr-screenshots
- project-conventions
- ralph-two-pass-scan
- reflect
- release-process
- reskill
- reviewer-protocol
- secret-handling
- session-recovery
- squad
- squad-commands
- squad-conventions
- squad-help
- squad-version-check
- test-discipline
- tiered-memory
- versioning-policy
- windows-compatibility
- {skill-name}
- agent-collaboration
- agent-conduct
- architectural-proposals
- ci-validation-gates
- client-compatibility
- coordinator-init-mode
- coordinator-response-mode
- coordinator-source-of-truth
- cross-machine-coordination
- cross-squad
- cross-squad-communication
- distributed-mesh
- docs-standards
- e2e-template-testing
- economy-mode
- error-recovery
- external-comms
- fact-checking
- gh-auth-isolation
- gh-aw-enlistment
- git-workflow
- github-multi-account
- history-hygiene
- humanizer
- init-mode
- iterative-retrieval
- model-selection
- nap
- notification-routing
- pr-review-response
- pr-screenshots
- project-conventions
- ralph-two-pass-scan
- reflect
- release-process
- reskill
- reviewer-protocol
- secret-handling
- session-recovery
- squad
- squad-commands
- squad-conventions
- squad-help
- squad-version-check
- test-discipline
- tiered-memory
- versioning-policy
- windows-compatibility
- {skill-name}
- agent-collaboration
- agent-conduct
- architectural-proposals
- ci-validation-gates
- client-compatibility
- coordinator-init-mode
- coordinator-response-mode
- coordinator-source-of-truth
- cross-machine-coordination
- cross-squad
- cross-squad-communication
- distributed-mesh
- docs-standards
- e2e-template-testing
- economy-mode
- error-recovery
- external-comms
- fact-checking
- gh-auth-isolation
- gh-aw-enlistment
- git-workflow
- github-multi-account
- history-hygiene
- humanizer
- init-mode
- iterative-retrieval
- model-selection
- nap
- notification-routing
- pr-review-response
- pr-screenshots
- project-conventions
- ralph-two-pass-scan
- reflect
- release-process
- reskill
- reviewer-protocol
- secret-handling
- session-recovery
- squad
- squad-commands
- squad-conventions
- squad-help
- squad-version-check
- test-discipline
- tiered-memory
- versioning-policy
- windows-compatibility
- My Skill
- gh-aw-enlistment
Repository facts
- License
- MIT
- Created
- Feb 6, 2026
- Last pushed
- Nov 6, 2026
- Forks
- 511
- Open issues
- 75
- Project activity
- Active
Rank over the last 12 weeks
Now #373 · best #278 · down 53 in 30 days
- #365 on Nov 2
- #366 on Nov 3
- #367 on Nov 4
- #368 on Nov 5
- #370 on Nov 6
- #373 since Nov 7
69 earlier periods
- #263 on Aug 2
- #270 on Aug 8
- #269 on Aug 9
- #277 on Aug 10
- #273 on Aug 11
- #274 on Aug 12
- #278 on Aug 13
- #276 on Aug 14
- #275 on Aug 15
- #280 on Aug 16
- #278 on Aug 17
- #287 on Aug 18
- #286 on Aug 19
- #283 on Aug 20
- #285 on Aug 21
- #279 on Aug 22
- #289 on Aug 24
- #290 from Aug 25 to Aug 26
- #287 on Aug 27
- #289 on Aug 28
- #288 on Aug 29
- #284 on Aug 30
- #291 on Aug 31
- #294 from Sep 1 to Sep 2
- #302 on Sep 3
- #294 on Sep 4
- #293 on Sep 5
- #303 on Sep 7
- #302 on Sep 8
- #303 on Sep 9
- #304 on Sep 10
- #301 on Sep 11
- #306 on Sep 14
- #307 on Sep 15
- #305 on Sep 16
- #303 from Sep 17 to Sep 18
- #299 on Sep 19
- #307 from Sep 20 to Sep 21
- #312 on Sep 22
- #310 on Sep 23
- #309 on Sep 24
- #307 on Sep 25
- #305 on Sep 26
- #314 on Sep 28
- #315 on Sep 29
- #312 on Sep 30
- #311 on Oct 1
- #315 on Oct 2
- #308 on Oct 4
- #314 on Oct 6
- #317 on Oct 7
- #320 on Oct 8
- #325 on Oct 9
- #316 on Oct 10
- #322 on Oct 11
- #331 on Oct 12
- #330 from Oct 13 to Oct 14
- #328 on Oct 15
- #331 on Oct 16
- #338 on Oct 19
- #339 from Oct 20 to Oct 21
- #342 on Oct 22
- #340 on Oct 23
- #338 on Oct 24
- #345 on Oct 25
- #357 on Oct 28
- #359 on Oct 29
- #362 on Oct 30
- #363 from Oct 31 to Nov 1
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 3, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 695 notes
.copilot/skills/archival-integrity/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/archival-integrity/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/archival-integrity/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/distributed-mesh/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/distributed-mesh/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/distributed-mesh/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/distributed-mesh/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/git-workflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/github-multi-account/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/github-multi-account/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/github-multi-account/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.copilot/skills/history-hygiene/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 33 dangerous shell commandsin 20 files: security-review/SKILL.md:77 windows-compatibility/SKILL.md:111 gh-aw-enlistment/SKILL.md:197 and 17 more
- 338 network referencesin 94 files: gh-auth-isolation/SKILL.md:56 gh-auth-isolation/SKILL.md:56 gh-auth-isolation/SKILL.md:56 and 91 more
- 4 obfuscation signalsin 4 files: e2e-template-testing/SKILL.md:309 e2e-template-testing/SKILL.md:309 e2e-template-testing/SKILL.md:309 and 1 more
- 1 instruction override signalin 1 file: model-selection/SKILL.md:56
- 44 executable scriptsin 29 files: scripts/bump-build.mjs:1 scripts/fix-crlf-worktree.mjs:1 scripts/capture-pr-screenshots.mjs:1 and 26 more
- 102 references to secretsin 23 files: gh-auth-isolation/SKILL.md:40 gh-auth-isolation/SKILL.md:40 gh-auth-isolation/SKILL.md:40 and 20 more
- No requests to vote in Hot
Show all 522 signals by file
.copilot/skills/ci-validation-gates/SKILL.md
- Network accessline 70
- Pin downloaded scripts and binaries to an immutable release or commit; never execute a moving branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Network accessline 71
branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Download to a named file before execution when practical, then remove it after a successful - Network accessline 72
- Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts.
.copilot/skills/distributed-mesh/SKILL.md
- Network accessline 59
1. SYNC: git pull (Zone 2) + curl (Zone 3) — materialize remote state - Network accessline 80
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 81
"zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 82
"source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 107
- **Phase 0:** Convention only — document zones, agree on mesh.json fields, manually run `git pull`/`git push`. Zero new code. - **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published - Network accessline 108
- **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears. - **Phase 3:** Never. No MCP federation, A2A, service discovery, message - Network accessline 109
- **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears.
.copilot/skills/distributed-mesh/mesh.json.example
- Network accessline 23
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", - Network accessline 24
"zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 25
"source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md",
.copilot/skills/distributed-mesh/sync-mesh.ps1
- Executable scriptline 1
Executable file
.copilot/skills/distributed-mesh/sync-mesh.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 7
# Usage: ./sync-mesh.sh [path-to-mesh.json] # ./sync-mesh.sh --init [path-to-mesh.json] # Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 9
# Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 100
eval curl --silent --fail $auth_flag "$source" -o "$target/SUMMARY.md" 2>/dev/null \
.copilot/skills/git-workflow/SKILL.md
- Dangerous shell operationline 131
If a worktree was deleted manually (rm -rf), `git worktree prune` recovers the state.
.copilot/skills/github-multi-account/SKILL.md
- Network accessline 3
description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically. confidence: high source: https://github.co - Network accessline 4
confidence: high source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account author: tamirdresher - Network accessline 5
source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
.copilot/skills/model-selection/SKILL.md
- Instruction override signalline 56
If a spawn fails because the selected model is unavailable (plan restriction, org policy, rate limit, deprecation, or any other reason), silently retry with the next model in the chain. Do NOT tell the user about fallback attempts. Maximum
.copilot/skills/secret-handling/SKILL.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 44
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Access to secretsline 56
[Credential-related source omitted] - Access to secretsline 99
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 126
[Credential-related source omitted]
.copilot/skills/security-review/SKILL.md
- Network accessline 36
**Watch for disguised secrets:** - Base64-encoded strings that decode to tokens - URLs with embedded credentials (`https://user:token@host`) - Network accessline 37
- Base64-encoded strings that decode to tokens - URLs with embedded credentials (`https://user:token@host`) - Test fixtures that contain real credentials from copy-paste - Network accessline 38
- URLs with embedded credentials (`https://user:token@host`) - Dangerous shell operationline 77
- ✅ Check for template literals in shell commands — `exec(\`git commit -m "${userMessage}"\`)` is injectable - Dangerous shell operationline 153
exec(`git commit -m "${commitMessage}"`) - Dangerous shell operationline 156
A message containing "; rm -rf /" would execute arbitrary commands. - Network accessline 213
[Credential-related source omitted] - Access to secretsline 213
[Credential-related source omitted] - Access to secretsline 214
[Credential-related source omitted] - Network accessline 215
[Credential-related source omitted] - Access to secretsline 217
[Credential-related source omitted] - Dangerous shell operationline 257
- ❌ Approving PRs with `exec()` and user-controlled input ("it's just a commit message")
.copilot/skills/windows-compatibility/SKILL.md
.squad-templates/cooperative-rate-limiting.md
- Network accessline 216
See [keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) for a complete implementation. - Network accessline 227
- [Circuit Breaker Template](ralph-circuit-breaker.md) — Foundation patterns - [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 228
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 229
- [KEDA Copilot Scaler](https://github.com/tamirdresher/keda-copilot-scaler) — Custom KEDA external scaler
.squad-templates/keda-scaler.md
- Network accessline 7
When running Squad on Kubernetes, agent pods sit idle when no work exists. [KEDA](https://keda.sh) (Kubernetes Event-Driven Autoscaler) solves this for queue-based workloads, but GitHub Issues isn't a native KEDA trigger. - Network accessline 34
kubectl apply -k https://github.com/tamirdresher/keda-copilot-scaler/deploy/kustomize - Network accessline 151
- **Repository:** [tamirdresher/keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) - Network accessline 161
- [KEDA External Scalers](https://keda.sh/docs/latest/concepts/external-scalers/) — KEDA documentation - Network accessline 162
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Full Kubernetes deployment example
.squad-templates/machine-capabilities.md
- Network accessline 75
A DaemonSet can run capability discovery on each node and maintain labels automatically. See the [squad-on-aks](https://github.com/tamirdresher/squad-on-aks) project for a complete Kubernetes deployment example.
.squad-templates/mcp-config.md
- Access to secretsline 23
[Credential-related source omitted] - Access to secretsline 24
[Credential-related source omitted] - Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 58
[Credential-related source omitted] - Network accessline 84
- **GitHub MCP requires a separate token** from the `gh` CLI auth. Generate at https://github.com/settings/tokens - Network accessline 85
- **Trello requires API key + token** from https://trello.com/power-ups/admin - Network accessline 87
- **Aspire uses the dashboard URL** — typically `http://localhost:18888` during local dev
.squad-templates/model-selection-reference.md
- Network accessline 131
Source of truth for public availability: the GitHub Copilot supported-models table (https://docs.github.com/en/copilot/reference/ai-models/supported-models), verified 2026-10-14. Only public GA models are listed; internal and preview models
.squad-templates/package.json
- Executable scriptline 1
Executable file
.squad-templates/ralph-triage.js
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 401
* Order: GITHUB_API_URL (set by Actions on both github.com and GHE runners) * > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 402
* > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). */ - Network accessline 403
* > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 412
return 'https://api.github.com'; - Access to secretsline 492
[Credential-related source omitted]
.squad-templates/scripts/notes/fetch.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
.squad-templates/scripts/notes/write-note.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
.squad-templates/session-init-reference.md
- Network accessline 165
- **Fallback chain:** - `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - Network accessline 166
- `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - No releases found → "No release notes found for this version range." - Network accessline 167
https://github.com/bradygaster/squad/releases"
.squad-templates/skills/ci-validation-gates/SKILL.md
- Network accessline 70
- Pin downloaded scripts and binaries to an immutable release or commit; never execute a moving branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Network accessline 71
branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Download to a named file before execution when practical, then remove it after a successful - Network accessline 72
- Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts.
.squad-templates/skills/distributed-mesh/SKILL.md
- Network accessline 59
1. SYNC: git pull (Zone 2) + curl (Zone 3) — materialize remote state - Network accessline 80
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 81
"zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 82
"source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 107
- **Phase 0:** Convention only — document zones, agree on mesh.json fields, manually run `git pull`/`git push`. Zero new code. - **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published - Network accessline 108
- **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears. - **Phase 3:** Never. No MCP federation, A2A, service discovery, message - Network accessline 109
- **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears.
.squad-templates/skills/distributed-mesh/mesh.json.example
- Network accessline 23
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", - Network accessline 24
"zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 25
"source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md",
.squad-templates/skills/distributed-mesh/sync-mesh.ps1
- Executable scriptline 1
Executable file
.squad-templates/skills/distributed-mesh/sync-mesh.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 7
# Usage: ./sync-mesh.sh [path-to-mesh.json] # ./sync-mesh.sh --init [path-to-mesh.json] # Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 9
# Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 100
eval curl --silent --fail $auth_flag "$source" -o "$target/SUMMARY.md" 2>/dev/null \
.squad-templates/skills/e2e-template-testing/SKILL.md
- Possible obfuscationline 309
character (`<U+FEFF>`) at the start of the comment.
.squad-templates/skills/external-comms/SKILL.md
.squad-templates/skills/gh-auth-isolation/SKILL.md
- Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 54
[Credential-related source omitted] - Network accessline 56
# 2. Push using token-authenticated HTTPS git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Network accessline 57
git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 73
[Credential-related source omitted] - Network accessline 131
[Credential-related source omitted] - Access to secretsline 131
[Credential-related source omitted] - Network accessline 132
git remote set-url origin https://personaluser:$token@github.com/personaluser/personaluser.github.io.git - Network accessline 135
# Clean up — don't leave token in remote URL git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 136
git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 160
# BAD: Hardcoding tokens in scripts git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 161
git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 170
[Credential-related source omitted] - Access to secretsline 171
[Credential-related source omitted] - Network accessline 172
git push https://personaluser:$token@github.com/personaluser/repo.git main - Access to secretsline 177
[Credential-related source omitted]
.squad-templates/skills/gh-aw-enlistment/SKILL.md
- Network accessline 26
it: each step below is a gate with explicit success evidence and **STOP conditions**. The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide an - Network accessline 27
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this skill ever disagree, the guide wins — re-read it before proceeding. - Network accessline 28
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this - Dangerous shell operationline 197
rm -f .github/skills/agentic-workflows/SKILL.md - Dangerous shell operationline 502
rm -f .github/skills/agentic-workflows/SKILL.md
.squad-templates/skills/git-workflow/SKILL.md
- Dangerous shell operationline 133
If a worktree was deleted manually (rm -rf), `git worktree prune` recovers the state.
.squad-templates/skills/github-multi-account/SKILL.md
- Network accessline 3
description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically. confidence: high source: https://github.co - Network accessline 4
confidence: high source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account author: tamirdresher - Network accessline 5
source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
.squad-templates/skills/humanizer/SKILL.md
- Network accessline 102
- ❌ Excessive emoji: More than 1-2 emoji per response - ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - Network accessline 103
- ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - ❌ Open-ended info requests: "Can you provide more information?" without specifying wha - Network accessline 104
- ❌ Link-dumping: Pasting URLs without context ("See: https://...")
.squad-templates/skills/model-selection/SKILL.md
- Network accessline 146
Chains list only public GA models (https://docs.github.com/en/copilot/reference/ai-models/supported-models, verified 2026-10-14). When the runtime supplies a live model list, it decides availability.
.squad-templates/skills/pr-screenshots/SKILL.md
- Network accessline 35
const screenshots = [ { url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 36
{ url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 37
{ url: 'http://localhost:4321/path/to/detail', name: 'feature-detail' }, - Network accessline 64
# Build raw URLs $base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 65
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 77
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots"
.squad-templates/skills/retro-enforcement/SKILL.md
- Network accessline 82
# Check for Issue references (e.g., #1478, https://github.com/.../issues/1478)
.squad-templates/skills/secret-handling/SKILL.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 44
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Access to secretsline 56
[Credential-related source omitted] - Access to secretsline 99
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 126
[Credential-related source omitted]
.squad-templates/skills/tiered-memory/SKILL.md
- Network accessline 11
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, - Network accessline 119
- Tracking issue: [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — installation gap + runtime status - Network accessline 120
- Original design spike: [bradygaster/squad#686](https://github.com/bradygaster/squad/issues/686) — tiered memory implementation plan - Network accessline 121
- Related: [bradygaster/squad#600](https://github.com/bradygaster/squad/issues/600) — context payload growth - Network accessline 218
- Cold adds a summary; only include when history is relevant - Wiki adds variable size; only include specific relevant docs - Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until t - Network accessline 220
- Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until those changes land, this skill is design-only and agents continue to load full history.md + decisions.md on every spawn
.squad-templates/skills/windows-compatibility/SKILL.md
.squad-templates/squad.agent.md
- Network accessline 335
<!-- Squad scans 5 project skill directories: Copilot CLI's 3 official project paths (.github/skills/, .claude/skills/, .agents/skills/) per https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-skills — plus Squad's - Network accessline 839
1. Run `gh --version`. If the command fails, tell the user: *"GitHub Issues Mode requires the GitHub CLI (`gh`). Install it from https://cli.github.com/ and run `gh auth login`."*
.squad-templates/workflows/squad-issue-assign.yml
- Access to secretsline 30
[Credential-related source omitted]
.squad-templates/workflows/squad-promote.yml
- Dangerous shell operationline 48
git rm -rf --cached --ignore-unmatch \
.squad/templates/mcp-config.md
- Access to secretsline 22
[Credential-related source omitted] - Access to secretsline 23
[Credential-related source omitted] - Access to secretsline 39
[Credential-related source omitted] - Access to secretsline 57
[Credential-related source omitted] - Network accessline 83
- **GitHub MCP requires a separate token** from the `gh` CLI auth. Generate at https://github.com/settings/tokens - Network accessline 84
- **Trello requires API key + token** from https://trello.com/power-ups/admin - Network accessline 86
- **Aspire uses the dashboard URL** — typically `http://localhost:18888` during local dev
CONTRIBUTING.md
- Network accessline 65
Go to https://github.com/bradygaster/squad and click "Fork" to create your own copy. - Network accessline 65
Go to https://github.com/bradygaster/squad and click "Fork" to create your own copy. - Network accessline 319
Squad uses [@changesets/cli](https://github.com/changesets/changesets) for independent package versioning. - Network accessline 319
Squad uses [@changesets/cli](https://github.com/changesets/changesets) for independent package versioning.
packages/squad-cli/templates/cooperative-rate-limiting.md
- Network accessline 216
See [keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) for a complete implementation. - Network accessline 227
- [Circuit Breaker Template](ralph-circuit-breaker.md) — Foundation patterns - [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 228
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 229
- [KEDA Copilot Scaler](https://github.com/tamirdresher/keda-copilot-scaler) — Custom KEDA external scaler
packages/squad-cli/templates/keda-scaler.md
- Network accessline 7
When running Squad on Kubernetes, agent pods sit idle when no work exists. [KEDA](https://keda.sh) (Kubernetes Event-Driven Autoscaler) solves this for queue-based workloads, but GitHub Issues isn't a native KEDA trigger. - Network accessline 34
kubectl apply -k https://github.com/tamirdresher/keda-copilot-scaler/deploy/kustomize - Network accessline 151
- **Repository:** [tamirdresher/keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) - Network accessline 161
- [KEDA External Scalers](https://keda.sh/docs/latest/concepts/external-scalers/) — KEDA documentation - Network accessline 162
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Full Kubernetes deployment example
packages/squad-cli/templates/machine-capabilities.md
- Network accessline 75
A DaemonSet can run capability discovery on each node and maintain labels automatically. See the [squad-on-aks](https://github.com/tamirdresher/squad-on-aks) project for a complete Kubernetes deployment example.
packages/squad-cli/templates/mcp-config.md
- Access to secretsline 23
[Credential-related source omitted] - Access to secretsline 24
[Credential-related source omitted] - Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 58
[Credential-related source omitted] - Network accessline 84
- **GitHub MCP requires a separate token** from the `gh` CLI auth. Generate at https://github.com/settings/tokens - Network accessline 85
- **Trello requires API key + token** from https://trello.com/power-ups/admin - Network accessline 87
- **Aspire uses the dashboard URL** — typically `http://localhost:18888` during local dev
packages/squad-cli/templates/model-selection-reference.md
- Network accessline 131
Source of truth for public availability: the GitHub Copilot supported-models table (https://docs.github.com/en/copilot/reference/ai-models/supported-models), verified 2026-10-14. Only public GA models are listed; internal and preview models
packages/squad-cli/templates/package.json
- Executable scriptline 1
Executable file
packages/squad-cli/templates/ralph-triage.js
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 401
* Order: GITHUB_API_URL (set by Actions on both github.com and GHE runners) * > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 402
* > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). */ - Network accessline 403
* > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 412
return 'https://api.github.com'; - Access to secretsline 492
[Credential-related source omitted]
packages/squad-cli/templates/scripts/notes/fetch.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
packages/squad-cli/templates/scripts/notes/write-note.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
packages/squad-cli/templates/session-init-reference.md
- Network accessline 165
- **Fallback chain:** - `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - Network accessline 166
- `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - No releases found → "No release notes found for this version range." - Network accessline 167
https://github.com/bradygaster/squad/releases"
packages/squad-cli/templates/skills/ci-validation-gates/SKILL.md
- Network accessline 70
- Pin downloaded scripts and binaries to an immutable release or commit; never execute a moving branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Network accessline 71
branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Download to a named file before execution when practical, then remove it after a successful - Network accessline 72
- Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts.
packages/squad-cli/templates/skills/distributed-mesh/SKILL.md
- Network accessline 59
1. SYNC: git pull (Zone 2) + curl (Zone 3) — materialize remote state - Network accessline 80
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 81
"zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 82
"source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 107
- **Phase 0:** Convention only — document zones, agree on mesh.json fields, manually run `git pull`/`git push`. Zero new code. - **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published - Network accessline 108
- **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears. - **Phase 3:** Never. No MCP federation, A2A, service discovery, message - Network accessline 109
- **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears.
packages/squad-cli/templates/skills/distributed-mesh/mesh.json.example
- Network accessline 23
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", - Network accessline 24
"zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 25
"source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md",
packages/squad-cli/templates/skills/distributed-mesh/sync-mesh.ps1
- Executable scriptline 1
Executable file
packages/squad-cli/templates/skills/distributed-mesh/sync-mesh.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 7
# Usage: ./sync-mesh.sh [path-to-mesh.json] # ./sync-mesh.sh --init [path-to-mesh.json] # Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 9
# Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 100
eval curl --silent --fail $auth_flag "$source" -o "$target/SUMMARY.md" 2>/dev/null \
packages/squad-cli/templates/skills/e2e-template-testing/SKILL.md
- Possible obfuscationline 309
character (`<U+FEFF>`) at the start of the comment.
packages/squad-cli/templates/skills/external-comms/SKILL.md
packages/squad-cli/templates/skills/gh-auth-isolation/SKILL.md
- Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 54
[Credential-related source omitted] - Network accessline 56
# 2. Push using token-authenticated HTTPS git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Network accessline 57
git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 73
[Credential-related source omitted] - Network accessline 131
[Credential-related source omitted] - Access to secretsline 131
[Credential-related source omitted] - Network accessline 132
git remote set-url origin https://personaluser:$token@github.com/personaluser/personaluser.github.io.git - Network accessline 135
# Clean up — don't leave token in remote URL git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 136
git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 160
# BAD: Hardcoding tokens in scripts git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 161
git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 170
[Credential-related source omitted] - Access to secretsline 171
[Credential-related source omitted] - Network accessline 172
git push https://personaluser:$token@github.com/personaluser/repo.git main - Access to secretsline 177
[Credential-related source omitted]
packages/squad-cli/templates/skills/gh-aw-enlistment/SKILL.md
- Network accessline 26
it: each step below is a gate with explicit success evidence and **STOP conditions**. The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide an - Network accessline 27
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this skill ever disagree, the guide wins — re-read it before proceeding. - Network accessline 28
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this - Dangerous shell operationline 197
rm -f .github/skills/agentic-workflows/SKILL.md - Dangerous shell operationline 502
rm -f .github/skills/agentic-workflows/SKILL.md
packages/squad-cli/templates/skills/git-workflow/SKILL.md
- Dangerous shell operationline 133
If a worktree was deleted manually (rm -rf), `git worktree prune` recovers the state.
packages/squad-cli/templates/skills/github-multi-account/SKILL.md
- Network accessline 3
description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically. confidence: high source: https://github.co - Network accessline 4
confidence: high source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account author: tamirdresher - Network accessline 5
source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
packages/squad-cli/templates/skills/humanizer/SKILL.md
- Network accessline 102
- ❌ Excessive emoji: More than 1-2 emoji per response - ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - Network accessline 103
- ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - ❌ Open-ended info requests: "Can you provide more information?" without specifying wha - Network accessline 104
- ❌ Link-dumping: Pasting URLs without context ("See: https://...")
packages/squad-cli/templates/skills/model-selection/SKILL.md
- Network accessline 146
Chains list only public GA models (https://docs.github.com/en/copilot/reference/ai-models/supported-models, verified 2026-10-14). When the runtime supplies a live model list, it decides availability.
packages/squad-cli/templates/skills/pr-screenshots/SKILL.md
- Network accessline 35
const screenshots = [ { url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 36
{ url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 37
{ url: 'http://localhost:4321/path/to/detail', name: 'feature-detail' }, - Network accessline 64
# Build raw URLs $base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 65
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 77
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots"
packages/squad-cli/templates/skills/retro-enforcement/SKILL.md
- Network accessline 82
# Check for Issue references (e.g., #1478, https://github.com/.../issues/1478)
packages/squad-cli/templates/skills/secret-handling/SKILL.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 44
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Access to secretsline 56
[Credential-related source omitted] - Access to secretsline 99
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 126
[Credential-related source omitted]
packages/squad-cli/templates/skills/tiered-memory/SKILL.md
- Network accessline 11
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, - Network accessline 119
- Tracking issue: [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — installation gap + runtime status - Network accessline 120
- Original design spike: [bradygaster/squad#686](https://github.com/bradygaster/squad/issues/686) — tiered memory implementation plan - Network accessline 121
- Related: [bradygaster/squad#600](https://github.com/bradygaster/squad/issues/600) — context payload growth - Network accessline 218
- Cold adds a summary; only include when history is relevant - Wiki adds variable size; only include specific relevant docs - Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until t - Network accessline 220
- Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until those changes land, this skill is design-only and agents continue to load full history.md + decisions.md on every spawn
packages/squad-cli/templates/skills/windows-compatibility/SKILL.md
packages/squad-cli/templates/squad.agent.md.template
- Network accessline 335
<!-- Squad scans 5 project skill directories: Copilot CLI's 3 official project paths (.github/skills/, .claude/skills/, .agents/skills/) per https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-skills — plus Squad's - Network accessline 839
1. Run `gh --version`. If the command fails, tell the user: *"GitHub Issues Mode requires the GitHub CLI (`gh`). Install it from https://cli.github.com/ and run `gh auth login`."*
packages/squad-cli/templates/workflows/squad-issue-assign.yml
- Access to secretsline 30
[Credential-related source omitted]
packages/squad-cli/templates/workflows/squad-promote.yml
- Dangerous shell operationline 48
git rm -rf --cached --ignore-unmatch \
packages/squad-sdk/templates/cooperative-rate-limiting.md
- Network accessline 216
See [keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) for a complete implementation. - Network accessline 227
- [Circuit Breaker Template](ralph-circuit-breaker.md) — Foundation patterns - [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 228
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 229
- [KEDA Copilot Scaler](https://github.com/tamirdresher/keda-copilot-scaler) — Custom KEDA external scaler
packages/squad-sdk/templates/keda-scaler.md
- Network accessline 7
When running Squad on Kubernetes, agent pods sit idle when no work exists. [KEDA](https://keda.sh) (Kubernetes Event-Driven Autoscaler) solves this for queue-based workloads, but GitHub Issues isn't a native KEDA trigger. - Network accessline 34
kubectl apply -k https://github.com/tamirdresher/keda-copilot-scaler/deploy/kustomize - Network accessline 151
- **Repository:** [tamirdresher/keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) - Network accessline 161
- [KEDA External Scalers](https://keda.sh/docs/latest/concepts/external-scalers/) — KEDA documentation - Network accessline 162
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Full Kubernetes deployment example
packages/squad-sdk/templates/machine-capabilities.md
- Network accessline 75
A DaemonSet can run capability discovery on each node and maintain labels automatically. See the [squad-on-aks](https://github.com/tamirdresher/squad-on-aks) project for a complete Kubernetes deployment example.
packages/squad-sdk/templates/mcp-config.md
- Access to secretsline 23
[Credential-related source omitted] - Access to secretsline 24
[Credential-related source omitted] - Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 58
[Credential-related source omitted] - Network accessline 84
- **GitHub MCP requires a separate token** from the `gh` CLI auth. Generate at https://github.com/settings/tokens - Network accessline 85
- **Trello requires API key + token** from https://trello.com/power-ups/admin - Network accessline 87
- **Aspire uses the dashboard URL** — typically `http://localhost:18888` during local dev
packages/squad-sdk/templates/model-selection-reference.md
- Network accessline 131
Source of truth for public availability: the GitHub Copilot supported-models table (https://docs.github.com/en/copilot/reference/ai-models/supported-models), verified 2026-10-14. Only public GA models are listed; internal and preview models
packages/squad-sdk/templates/package.json
- Executable scriptline 1
Executable file
packages/squad-sdk/templates/ralph-triage.js
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 401
* Order: GITHUB_API_URL (set by Actions on both github.com and GHE runners) * > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 402
* > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). */ - Network accessline 403
* > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 412
return 'https://api.github.com'; - Access to secretsline 492
[Credential-related source omitted]
packages/squad-sdk/templates/scripts/notes/fetch.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
packages/squad-sdk/templates/scripts/notes/write-note.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
packages/squad-sdk/templates/session-init-reference.md
- Network accessline 165
- **Fallback chain:** - `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - Network accessline 166
- `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - No releases found → "No release notes found for this version range." - Network accessline 167
https://github.com/bradygaster/squad/releases"
packages/squad-sdk/templates/skills/ci-validation-gates/SKILL.md
- Network accessline 70
- Pin downloaded scripts and binaries to an immutable release or commit; never execute a moving branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Network accessline 71
branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Download to a named file before execution when practical, then remove it after a successful - Network accessline 72
- Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts.
packages/squad-sdk/templates/skills/distributed-mesh/SKILL.md
- Network accessline 59
1. SYNC: git pull (Zone 2) + curl (Zone 3) — materialize remote state - Network accessline 80
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 81
"zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 82
"source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 107
- **Phase 0:** Convention only — document zones, agree on mesh.json fields, manually run `git pull`/`git push`. Zero new code. - **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published - Network accessline 108
- **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears. - **Phase 3:** Never. No MCP federation, A2A, service discovery, message - Network accessline 109
- **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears.
packages/squad-sdk/templates/skills/distributed-mesh/mesh.json.example
- Network accessline 23
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", - Network accessline 24
"zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 25
"source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md",
packages/squad-sdk/templates/skills/distributed-mesh/sync-mesh.ps1
- Executable scriptline 1
Executable file
packages/squad-sdk/templates/skills/distributed-mesh/sync-mesh.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 7
# Usage: ./sync-mesh.sh [path-to-mesh.json] # ./sync-mesh.sh --init [path-to-mesh.json] # Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 9
# Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 100
eval curl --silent --fail $auth_flag "$source" -o "$target/SUMMARY.md" 2>/dev/null \
packages/squad-sdk/templates/skills/e2e-template-testing/SKILL.md
- Possible obfuscationline 309
character (`<U+FEFF>`) at the start of the comment.
packages/squad-sdk/templates/skills/external-comms/SKILL.md
packages/squad-sdk/templates/skills/gh-auth-isolation/SKILL.md
- Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 54
[Credential-related source omitted] - Network accessline 56
# 2. Push using token-authenticated HTTPS git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Network accessline 57
git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 73
[Credential-related source omitted] - Network accessline 131
[Credential-related source omitted] - Access to secretsline 131
[Credential-related source omitted] - Network accessline 132
git remote set-url origin https://personaluser:$token@github.com/personaluser/personaluser.github.io.git - Network accessline 135
# Clean up — don't leave token in remote URL git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 136
git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 160
# BAD: Hardcoding tokens in scripts git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 161
git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 170
[Credential-related source omitted] - Access to secretsline 171
[Credential-related source omitted] - Network accessline 172
git push https://personaluser:$token@github.com/personaluser/repo.git main - Access to secretsline 177
[Credential-related source omitted]
packages/squad-sdk/templates/skills/gh-aw-enlistment/SKILL.md
- Network accessline 26
it: each step below is a gate with explicit success evidence and **STOP conditions**. The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide an - Network accessline 27
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this skill ever disagree, the guide wins — re-read it before proceeding. - Network accessline 28
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this - Dangerous shell operationline 197
rm -f .github/skills/agentic-workflows/SKILL.md - Dangerous shell operationline 502
rm -f .github/skills/agentic-workflows/SKILL.md
packages/squad-sdk/templates/skills/git-workflow/SKILL.md
- Dangerous shell operationline 133
If a worktree was deleted manually (rm -rf), `git worktree prune` recovers the state.
packages/squad-sdk/templates/skills/github-multi-account/SKILL.md
- Network accessline 3
description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically. confidence: high source: https://github.co - Network accessline 4
confidence: high source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account author: tamirdresher - Network accessline 5
source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
packages/squad-sdk/templates/skills/humanizer/SKILL.md
- Network accessline 102
- ❌ Excessive emoji: More than 1-2 emoji per response - ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - Network accessline 103
- ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - ❌ Open-ended info requests: "Can you provide more information?" without specifying wha - Network accessline 104
- ❌ Link-dumping: Pasting URLs without context ("See: https://...")
packages/squad-sdk/templates/skills/model-selection/SKILL.md
- Network accessline 146
Chains list only public GA models (https://docs.github.com/en/copilot/reference/ai-models/supported-models, verified 2026-10-14). When the runtime supplies a live model list, it decides availability.
packages/squad-sdk/templates/skills/pr-screenshots/SKILL.md
- Network accessline 35
const screenshots = [ { url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 36
{ url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 37
{ url: 'http://localhost:4321/path/to/detail', name: 'feature-detail' }, - Network accessline 64
# Build raw URLs $base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 65
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 77
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots"
packages/squad-sdk/templates/skills/retro-enforcement/SKILL.md
- Network accessline 82
# Check for Issue references (e.g., #1478, https://github.com/.../issues/1478)
packages/squad-sdk/templates/skills/secret-handling/SKILL.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 44
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Access to secretsline 56
[Credential-related source omitted] - Access to secretsline 99
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 126
[Credential-related source omitted]
packages/squad-sdk/templates/skills/tiered-memory/SKILL.md
- Network accessline 11
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, - Network accessline 119
- Tracking issue: [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — installation gap + runtime status - Network accessline 120
- Original design spike: [bradygaster/squad#686](https://github.com/bradygaster/squad/issues/686) — tiered memory implementation plan - Network accessline 121
- Related: [bradygaster/squad#600](https://github.com/bradygaster/squad/issues/600) — context payload growth - Network accessline 218
- Cold adds a summary; only include when history is relevant - Wiki adds variable size; only include specific relevant docs - Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until t - Network accessline 220
- Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until those changes land, this skill is design-only and agents continue to load full history.md + decisions.md on every spawn
packages/squad-sdk/templates/skills/windows-compatibility/SKILL.md
packages/squad-sdk/templates/squad.agent.md.template
- Network accessline 335
<!-- Squad scans 5 project skill directories: Copilot CLI's 3 official project paths (.github/skills/, .claude/skills/, .agents/skills/) per https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-skills — plus Squad's - Network accessline 839
1. Run `gh --version`. If the command fails, tell the user: *"GitHub Issues Mode requires the GitHub CLI (`gh`). Install it from https://cli.github.com/ and run `gh auth login`."*
packages/squad-sdk/templates/workflows/squad-issue-assign.yml
- Access to secretsline 30
[Credential-related source omitted]
packages/squad-sdk/templates/workflows/squad-promote.yml
- Dangerous shell operationline 48
git rm -rf --cached --ignore-unmatch \
scripts/bump-build.mjs
- Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
#!/usr/bin/env node
scripts/capture-pr-screenshots.mjs
- Executable scriptline 1
Executable file - Executable scriptline 1
Executable file - Executable scriptline 1
Executable file - Executable scriptline 1
Executable file - Network accessline 4
const BASE = 'http://localhost:4321/squad/docs/reference/api'; - Network accessline 4
const BASE = 'http://localhost:4321/squad/docs/reference/api'; - Network accessline 4
const BASE = 'http://localhost:4321/squad/docs/reference/api'; - Network accessline 4
const BASE = 'http://localhost:4321/squad/docs/reference/api'; - Network accessline 21
const healthCheck = await fetch(BASE).catch(() => null); - Network accessline 21
const healthCheck = await fetch(BASE).catch(() => null); - Network accessline 21
const healthCheck = await fetch(BASE).catch(() => null); - Network accessline 21
const healthCheck = await fetch(BASE).catch(() => null);
scripts/fix-crlf-worktree.mjs
templates/cooperative-rate-limiting.md
- Network accessline 216
See [keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) for a complete implementation. - Network accessline 227
- [Circuit Breaker Template](ralph-circuit-breaker.md) — Foundation patterns - [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 228
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Production K8s deployment - Network accessline 229
- [KEDA Copilot Scaler](https://github.com/tamirdresher/keda-copilot-scaler) — Custom KEDA external scaler
templates/keda-scaler.md
- Network accessline 7
When running Squad on Kubernetes, agent pods sit idle when no work exists. [KEDA](https://keda.sh) (Kubernetes Event-Driven Autoscaler) solves this for queue-based workloads, but GitHub Issues isn't a native KEDA trigger. - Network accessline 34
kubectl apply -k https://github.com/tamirdresher/keda-copilot-scaler/deploy/kustomize - Network accessline 151
- **Repository:** [tamirdresher/keda-copilot-scaler](https://github.com/tamirdresher/keda-copilot-scaler) - Network accessline 161
- [KEDA External Scalers](https://keda.sh/docs/latest/concepts/external-scalers/) — KEDA documentation - Network accessline 162
- [Squad on AKS](https://github.com/tamirdresher/squad-on-aks) — Full Kubernetes deployment example
templates/machine-capabilities.md
- Network accessline 75
A DaemonSet can run capability discovery on each node and maintain labels automatically. See the [squad-on-aks](https://github.com/tamirdresher/squad-on-aks) project for a complete Kubernetes deployment example.
templates/mcp-config.md
- Access to secretsline 23
[Credential-related source omitted] - Access to secretsline 24
[Credential-related source omitted] - Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 58
[Credential-related source omitted] - Network accessline 84
- **GitHub MCP requires a separate token** from the `gh` CLI auth. Generate at https://github.com/settings/tokens - Network accessline 85
- **Trello requires API key + token** from https://trello.com/power-ups/admin - Network accessline 87
- **Aspire uses the dashboard URL** — typically `http://localhost:18888` during local dev
templates/model-selection-reference.md
- Network accessline 131
Source of truth for public availability: the GitHub Copilot supported-models table (https://docs.github.com/en/copilot/reference/ai-models/supported-models), verified 2026-10-14. Only public GA models are listed; internal and preview models
templates/package.json
- Executable scriptline 1
Executable file
templates/ralph-triage.js
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 401
* Order: GITHUB_API_URL (set by Actions on both github.com and GHE runners) * > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 402
* > GITHUB_SERVER_URL + /api/v3 (GHE without an explicit API URL) * > https://api.github.com (fallback for local/non-Actions runs). */ - Network accessline 403
* > https://api.github.com (fallback for local/non-Actions runs). - Network accessline 412
return 'https://api.github.com'; - Access to secretsline 492
[Credential-related source omitted]
templates/scripts/notes/fetch.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
templates/scripts/notes/write-note.ps1
- Executable scriptline 1
#!/usr/bin/env pwsh
templates/session-init-reference.md
- Network accessline 165
- **Fallback chain:** - `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - Network accessline 166
- `gh` not authenticated → "See full release notes at: https://github.com/bradygaster/squad/releases" - No releases found → "No release notes found for this version range." - Network accessline 167
https://github.com/bradygaster/squad/releases"
templates/skills/ci-validation-gates/SKILL.md
- Network accessline 70
- Pin downloaded scripts and binaries to an immutable release or commit; never execute a moving branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Network accessline 71
branch reference. - Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts. - Download to a named file before execution when practical, then remove it after a successful - Network accessline 72
- Use `curl -f` (normally `curl -fsSL`) so HTTP failures cannot be interpreted as scripts.
templates/skills/distributed-mesh/SKILL.md
- Network accessline 59
1. SYNC: git pull (Zone 2) + curl (Zone 3) — materialize remote state - Network accessline 80
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 81
"zone": "remote-opaque", "source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 82
"source": "https://partner.dev/squad-contracts/fraud/SUMMARY.md", - Network accessline 107
- **Phase 0:** Convention only — document zones, agree on mesh.json fields, manually run `git pull`/`git push`. Zero new code. - **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published - Network accessline 108
- **Phase 1:** Sync script (~30 lines bash or PowerShell) when manual sync gets tedious. - **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears. - **Phase 3:** Never. No MCP federation, A2A, service discovery, message - Network accessline 109
- **Phase 2:** Published contracts + curl fetch when a Zone 3 partner appears.
templates/skills/distributed-mesh/mesh.json.example
- Network accessline 23
"partner-fraud": { "zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", - Network accessline 24
"zone": "remote-opaque", "source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md", "sync_to": ".mesh/remotes/partner-fraud", - Network accessline 25
"source": "https://partner.example.com/squad-contracts/fraud/SUMMARY.md",
templates/skills/distributed-mesh/sync-mesh.ps1
- Executable scriptline 1
Executable file
templates/skills/distributed-mesh/sync-mesh.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 7
# Usage: ./sync-mesh.sh [path-to-mesh.json] # ./sync-mesh.sh --init [path-to-mesh.json] # Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 9
# Requires: jq (https://github.com/jqlang/jq), git, curl - Network accessline 100
eval curl --silent --fail $auth_flag "$source" -o "$target/SUMMARY.md" 2>/dev/null \
templates/skills/e2e-template-testing/SKILL.md
- Possible obfuscationline 309
character (`<U+FEFF>`) at the start of the comment.
templates/skills/external-comms/SKILL.md
templates/skills/gh-auth-isolation/SKILL.md
- Access to secretsline 40
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 54
[Credential-related source omitted] - Network accessline 56
# 2. Push using token-authenticated HTTPS git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Network accessline 57
git push https://personaluser:$token@github.com/personaluser/repo.git branch-name - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 73
[Credential-related source omitted] - Network accessline 131
[Credential-related source omitted] - Access to secretsline 131
[Credential-related source omitted] - Network accessline 132
git remote set-url origin https://personaluser:$token@github.com/personaluser/personaluser.github.io.git - Network accessline 135
# Clean up — don't leave token in remote URL git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 136
git remote set-url origin https://github.com/personaluser/personaluser.github.io.git - Network accessline 160
# BAD: Hardcoding tokens in scripts git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 161
git push https://personaluser:ghp_xxxxxxxxxxxx@github.com/personaluser/repo.git main - Network accessline 170
[Credential-related source omitted] - Access to secretsline 171
[Credential-related source omitted] - Network accessline 172
git push https://personaluser:$token@github.com/personaluser/repo.git main - Access to secretsline 177
[Credential-related source omitted]
templates/skills/gh-aw-enlistment/SKILL.md
- Network accessline 26
it: each step below is a gate with explicit success evidence and **STOP conditions**. The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide an - Network accessline 27
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this skill ever disagree, the guide wins — re-read it before proceeding. - Network accessline 28
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this - Dangerous shell operationline 197
rm -f .github/skills/agentic-workflows/SKILL.md - Dangerous shell operationline 502
rm -f .github/skills/agentic-workflows/SKILL.md
templates/skills/git-workflow/SKILL.md
- Dangerous shell operationline 133
If a worktree was deleted manually (rm -rf), `git worktree prune` recovers the state.
templates/skills/github-multi-account/SKILL.md
- Network accessline 3
description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically. confidence: high source: https://github.co - Network accessline 4
confidence: high source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account author: tamirdresher - Network accessline 5
source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
templates/skills/humanizer/SKILL.md
- Network accessline 102
- ❌ Excessive emoji: More than 1-2 emoji per response - ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - Network accessline 103
- ❌ Quoting profanity: Even when the original issue contains it, paraphrase instead - ❌ Link-dumping: Pasting URLs without context ("See: https://...") - ❌ Open-ended info requests: "Can you provide more information?" without specifying wha - Network accessline 104
- ❌ Link-dumping: Pasting URLs without context ("See: https://...")
templates/skills/model-selection/SKILL.md
- Network accessline 146
Chains list only public GA models (https://docs.github.com/en/copilot/reference/ai-models/supported-models, verified 2026-10-14). When the runtime supplies a live model list, it decides availability.
templates/skills/pr-screenshots/SKILL.md
- Network accessline 35
const screenshots = [ { url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 36
{ url: 'http://localhost:4321/path/to/page', name: 'feature-landing' }, - Network accessline 37
{ url: 'http://localhost:4321/path/to/detail', name: 'feature-detail' }, - Network accessline 64
# Build raw URLs $base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 65
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots" - Network accessline 77
$base = "https://raw.githubusercontent.com/{owner}/{repo}/screenshots-temp/screenshots"
templates/skills/retro-enforcement/SKILL.md
- Network accessline 82
# Check for Issue references (e.g., #1478, https://github.com/.../issues/1478)
templates/skills/secret-handling/SKILL.md
- Access to secretsline 36
[Credential-related source omitted] - Access to secretsline 44
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Access to secretsline 56
[Credential-related source omitted] - Access to secretsline 99
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 126
[Credential-related source omitted]
templates/skills/tiered-memory/SKILL.md
- Network accessline 11
> **Status (v0.10.0):** This skill describes a **design proposal**, not a shipped runtime. Skill files install via `squad init`/`upgrade`, but the underlying tier scaffolding (`.squad/memory/hot/`, `cold/`, `wiki/`), Scribe promotion logic, - Network accessline 119
- Tracking issue: [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — installation gap + runtime status - Network accessline 120
- Original design spike: [bradygaster/squad#686](https://github.com/bradygaster/squad/issues/686) — tiered memory implementation plan - Network accessline 121
- Related: [bradygaster/squad#600](https://github.com/bradygaster/squad/issues/600) — context payload growth - Network accessline 218
- Cold adds a summary; only include when history is relevant - Wiki adds variable size; only include specific relevant docs - Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until t - Network accessline 220
- Runtime backing is tracked in [bradygaster/squad#1264](https://github.com/bradygaster/squad/issues/1264) — until those changes land, this skill is design-only and agents continue to load full history.md + decisions.md on every spawn
templates/skills/windows-compatibility/SKILL.md
templates/squad.agent.md.template
- Network accessline 335
<!-- Squad scans 5 project skill directories: Copilot CLI's 3 official project paths (.github/skills/, .claude/skills/, .agents/skills/) per https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-skills — plus Squad's - Network accessline 839
1. Run `gh --version`. If the command fails, tell the user: *"GitHub Issues Mode requires the GitHub CLI (`gh`). Install it from https://cli.github.com/ and run `gh auth login`."*
templates/workflows/squad-issue-assign.yml
- Access to secretsline 30
[Credential-related source omitted]
templates/workflows/squad-promote.yml
- Dangerous shell operationline 48
git rm -rf --cached --ignore-unmatch \
workflows/skills/gh-aw-enlistment/SKILL.md
- Network accessline 26
it: each step below is a gate with explicit success evidence and **STOP conditions**. The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide an - Network accessline 27
The authoritative source is `docs/src/content/docs/guide/gh-aw.md` (mirrored at `https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this skill ever disagree, the guide wins — re-read it before proceeding. - Network accessline 28
`https://bradygaster.github.io/squad/docs/guide/gh-aw/`). If that guide and this - Dangerous shell operationline 197
rm -f .github/skills/agentic-workflows/SKILL.md - Dangerous shell operationline 502
rm -f .github/skills/agentic-workflows/SKILL.md
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Signal confidence: 493 facts
- Network access in prose of the SKILL.mdci-validation-gates/SKILL.md:70Low confidence
- Network access in prose of the SKILL.mdci-validation-gates/SKILL.md:71Low confidence
- Network access in prose of the SKILL.mdci-validation-gates/SKILL.md:72Low confidence
- Network access in code of the SKILL.mddistributed-mesh/SKILL.md:59High confidence
- Network access in code of the SKILL.mddistributed-mesh/SKILL.md:80High confidence
- Network access in code of the SKILL.mddistributed-mesh/SKILL.md:81High confidence
- Network access in code of the SKILL.mddistributed-mesh/SKILL.md:82High confidence
- Network access in prose of the SKILL.mddistributed-mesh/SKILL.md:107Low confidence
- Network access in prose of the SKILL.mddistributed-mesh/SKILL.md:108Low confidence
- Network access in prose of the SKILL.mddistributed-mesh/SKILL.md:109Low confidence
- Executable script in a comment of a file that runsdistributed-mesh/sync-mesh.ps1:1Low confidence
- Executable script in code of a file that runsdistributed-mesh/sync-mesh.sh:1High confidence
- Network access in a comment of a file that runsdistributed-mesh/sync-mesh.sh:7Low confidence
- Network access in a comment of a file that runsdistributed-mesh/sync-mesh.sh:9Low confidence
- Network access in code of a file that runsdistributed-mesh/sync-mesh.sh:100High confidence
- Dangerous shell operation in prose of the SKILL.mdgit-workflow/SKILL.md:131Low confidence
- Network access in the frontmatter of the SKILL.mdgithub-multi-account/SKILL.md:3Low confidence
- Network access in the frontmatter of the SKILL.mdgithub-multi-account/SKILL.md:4Low confidence
- Network access in the frontmatter of the SKILL.mdgithub-multi-account/SKILL.md:5Low confidence
- Instruction override signal in prose of the SKILL.mdmodel-selection/SKILL.md:56High confidence
- Access to secrets in prose of the SKILL.mdsecret-handling/SKILL.md:36Low confidence
- Access to secrets in prose of the SKILL.mdsecret-handling/SKILL.md:44Low confidence
- Access to secrets in prose of the SKILL.md: a variable, placeholder or call, not a stored valuesecret-handling/SKILL.md:53Low confidence
- Access to secrets in prose of the SKILL.md: a variable, placeholder or call, not a stored valuesecret-handling/SKILL.md:56Low confidence
- Access to secrets in code of the SKILL.mdsecret-handling/SKILL.md:99High confidence
- Access to secrets in code of the SKILL.mdsecret-handling/SKILL.md:120High confidence
- Access to secrets in code of the SKILL.md: a variable, placeholder or call, not a stored valuesecret-handling/SKILL.md:126Low confidence
- Network access in prose of the SKILL.mdsecurity-review/SKILL.md:36Low confidence
- Network access in prose of the SKILL.mdsecurity-review/SKILL.md:37Low confidence
- Network access in prose of the SKILL.mdsecurity-review/SKILL.md:38Low confidence
- Dangerous shell operation in prose of the SKILL.mdsecurity-review/SKILL.md:77Low confidence
- Dangerous shell operation in code of the SKILL.mdsecurity-review/SKILL.md:153High confidence
- Dangerous shell operation in code of the SKILL.mdsecurity-review/SKILL.md:156High confidence
- Network access in code of the SKILL.mdsecurity-review/SKILL.md:213High confidence
- Access to secrets in code of the SKILL.md: a variable, placeholder or call, not a stored valuesecurity-review/SKILL.md:213Low confidence
- Access to secrets in code of the SKILL.md: a variable, placeholder or call, not a stored valuesecurity-review/SKILL.md:214Low confidence
- Network access in code of the SKILL.mdsecurity-review/SKILL.md:215High confidence
- Access to secrets in code of the SKILL.mdsecurity-review/SKILL.md:217High confidence
- Dangerous shell operation in prose of the SKILL.mdsecurity-review/SKILL.md:257Low confidence
- Dangerous shell operation in code of the SKILL.mdwindows-compatibility/SKILL.md:111High confidence
- Dangerous shell operation in code of the SKILL.mdwindows-compatibility/SKILL.md:114High confidence
- Network access in prose of documentation.squad-templates/cooperative-rate-limiting.md:216Low confidence
- Network access in prose of documentation.squad-templates/cooperative-rate-limiting.md:227Low confidence
- Network access in prose of documentation.squad-templates/cooperative-rate-limiting.md:228Low confidence
- Network access in prose of documentation.squad-templates/cooperative-rate-limiting.md:229Low confidence
- Network access in prose of documentation.squad-templates/keda-scaler.md:7Low confidence
- Network access in code of documentation.squad-templates/keda-scaler.md:34Low confidence
- Network access in prose of documentation.squad-templates/keda-scaler.md:151Low confidence
- Network access in prose of documentation.squad-templates/keda-scaler.md:161Low confidence
- Network access in prose of documentation.squad-templates/keda-scaler.md:162Low confidence
and 443 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Frontmatter field user-invocable in .squad-templates/skills/squad/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md .copilot/skills/action-skill-contracts/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md .copilot/skills/action-skill-contracts/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md .copilot/skills/action-skill-contracts/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Native | Manifest .github/copilot-instructions.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md .copilot/skills/action-skill-contracts/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
Explore alternatives and comparisons
Add this badge to your README
It updates every night and links to this page.
[](https://hotskills.dev/skill/github-squad/)