Menu

#292 by GitHub stars

BuilderIO/skills

Skills for agents

Claude CodeEngineering

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

skills/agent-watchdog/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/an/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/efficient-fable/SKILL.md

Week of Oct 26, 2026 · +4 / −90 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

- 3 description: Use when running Claude Fable on codebase-heavy or token-heavy work and the user wants Fable to orchestrate research, coding, and testing while cheaper subagents do bounded heavy lifting.
+ 3 description: Alias for /efficient-orchestrator. Run only when the user explicitly invokes /efficient-fable; never self-activate.
+ 4 disable-model-invocation: true
- 6 # Efficient Fable
+ 7 # /efficient-fable
- 8 Use Claude Fable as the orchestrator, architect, synthesizer, and final judge.
- 9 Use cheaper subagents for token-heavy research, coding, testing, and
- 10 summarization that do not require Fable's full judgment.
- 11 
- 12 ## Where Fable Shines
- 13 
- 14 Reserve Fable for:
- 15 
- 16 - Decomposing ambiguous work into clean parallel slices.
- 17 - Architecture, product, and safety tradeoffs.
- 18 - Reading conflicting subagent reports and deciding what matters.
- 19 - Integrating partial implementations into one coherent plan.
- 20 - Final review, risk assessment, and user-facing synthesis.
- 21 
- 22 ## Delegation Pattern
- 23 
- 24 1. Name the expensive-token risk: large repo search, long logs, broad docs, or
- 25    repetitive edits.
- 26 2. Split independent work into subagents before reading everything yourself.
- 27 3. Use cheaper models for research scans, inventory, search summaries, narrow
- 28    bug hunts, browser/testing passes, test output reduction, and bounded code
- 29    edits.
- 30 4. Ask subagents for concise evidence: files, line references, commands run,
- 31    diffs, uncertainties, and stop conditions they hit.
- 32 5. Spend Fable tokens on the decision layer: compare results, resolve conflicts,
- 33    choose the implementation path, and review the final patch.
- 34 
- 35 Prefer parallel subagents when the slices do not depend on each other. Keep
- 36 blocking or highly coupled work local.
- 37 
- 38 ## Handoff Packets
- 39 
- 40 Write delegated prompts as if the subagent has no useful chat context. Include
- 41 only the context it needs:
- 42 
- 43 - The repo path and exact objective.
- 44 - The files, packages, or surfaces in scope and anything explicitly out of
- 45   scope.
- 46 - The evidence format to return: files, line refs, commands, diffs, failures,
- 47   screenshots, and uncertainty.
- 48 - The verification commands or browser flows to run, plus what success should
- 49   look like when that is knowable.
- 50 - Stop conditions: if the code does not match the prompt, a command fails after
- 51   a reasonable retry, or the task needs out-of-scope files, stop and report
- 52   instead of improvising.
- 53 
- 54 ## Vetting Delegated Work
- 55 
- 56 Treat subagent reports as leads, not facts. Before using a high-impact finding,
- 57 opening a PR, or telling the user the work is done, Fable should reopen the
- 58 important cited files, confirm the relevant line refs or failures, and review
- 59 the final diff against the task. Let lighter agents gather signal; keep
- 60 truth-judgment with Fable.
- 61 
- 62 ## Common Scenarios
- 63 
- 64 Treat these as soft defaults, not rigid rules:
- 65 
- 66 - Research: ask lighter agents to scan docs, prior art, APIs, and repo surfaces;
- 67   Fable decides what evidence changes the plan.
- 68 - Coding: give cheaper agents bounded edits or candidate patches; Fable owns
- 69   shared-file coordination, integration, and final review.
- 70 - Testing: have Fable suggest the validation direction and the scripts or
- 71   browser checks that matter. Let lighter agents run targeted tests, browser
- 72   flows, screenshots, and log reduction, then report exact commands, failures,
- 73   likely causes, and whether failures look flaky, environmental, or real.
- 74 - Debugging: use cheaper agents to cluster logs, reproduce issues, and try
- 75   small fixes; Fable decides which diagnosis is most trustworthy.
- 76 
- 77 If a task is tiny or the validation itself needs delicate judgment, keep it
- 78 with Fable.
- 79 
- 80 ## Diagram
- 81 
- 82 Use `assets/fable-orchestrator.excalidraw` when a visual explanation helps.
- 83 
- 84 ## Claims
- 85 
- 86 For codebase-heavy work, it is reasonable to describe this as up to 3-5x more
- 87 cost-efficient and 2-4x faster when independent research, coding, or testing
- 88 slices can run in parallel. Treat those as workload-dependent estimates, not
- 89 guarantees.
- 90 
- 91 Good launch copy:
- 92 
- 93 > Make Claude Fable more efficient by using cheaper subagents for token-heavy
- 94 > research, coding, and testing, saving Fable for judgment, architecture,
- 95 > synthesis, and final review.
+ 9 The user asked for /efficient-fable, an alias of /efficient-orchestrator. Follow the efficient-orchestrator skill exactly; install it alongside this alias.

Compare on GitHub

skills/efficient-frontier/SKILL.md

Week of Oct 26, 2026 · +4 / −71 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

- 3 description: >-
- 4   Apply the same orchestration as `/efficient-fable` to any high-cost frontier
- 5   model: delegate research, coding, and testing to cheaper subagents while
- 6   keeping planning, synthesis, and final review with the expensive model.
+ 3 description: Alias for /efficient-orchestrator. Run only when the user explicitly invokes /efficient-frontier; never self-activate.
+ 4 disable-model-invocation: true
- 9 # Efficient Frontier
+ 7 # /efficient-frontier
- 11 Use the expensive frontier model where its marginal judgment matters. Push
- 12 repeatable, bounded, or token-heavy work to cheaper/faster subagents.
- 13 
- 14 ## Workflow
- 15 
- 16 1. Identify the frontier-only decisions: architecture, prioritization,
- 17    ambiguity resolution, risk, synthesis, and final review.
- 18 2. Identify delegable work: research scans, repository inventory, search, docs
- 19    extraction, browser/testing passes, log reduction, test failure clustering,
- 20    narrow coding, and mechanical edits.
- 21 3. Spawn parallel subagents for independent slices with clear ownership,
- 22    bounded scope, verification gates, and expected evidence.
- 23 4. Require compact returns: findings, changed files, commands run, residual
- 24    risk, stop conditions hit, and anything the frontier model must decide.
- 25 5. Integrate and review centrally before presenting the result.
- 26 
- 27 ## Handoff Packets
- 28 
- 29 Write delegated prompts as self-contained packets. Assume the receiving agent
- 30 has not seen the conversation. Include the repo path, objective, scope,
- 31 out-of-scope areas, relevant files or search targets, expected return format,
- 32 verification commands, and stop conditions.
- 33 
- 34 Useful stop conditions:
- 35 
- 36 - The live code does not match the assumption in the handoff.
- 37 - A verification command fails twice after a reasonable fix or retry.
- 38 - The work appears to require files outside the assigned scope.
- 39 - The agent cannot produce concrete evidence for its claim.
- 40 
- 41 ## Review Loop
- 42 
- 43 Treat delegated output as evidence to inspect, not a verdict to forward. Reopen
- 44 important cited files, skim high-risk diffs, and rerun or spot-check the
- 45 verification that matters before claiming completion. If delegated agents
- 46 disagree, resolve the disagreement at the frontier-model layer.
- 47 
- 48 ## Common Scenarios
- 49 
- 50 Use these as soft suggestions:
- 51 
- 52 - Research: delegate broad repo scans, docs extraction, and source comparison;
- 53   the frontier model keeps the judgment about what matters.
- 54 - Coding: delegate bounded patches, refactors, or mechanical edits when file
- 55   ownership is clear; integrate and review centrally.
- 56 - Testing: let the frontier model choose the validation strategy and scripts,
- 57   then use cheaper agents to run unit checks, browser flows, screenshots, and
- 58   log reduction. Ask them to return exact commands, failures, likely causes, and
- 59   whether the signal looks flaky, environmental, or product-relevant.
- 60 - Debugging: send independent agents after separate theories, logs, or repro
- 61   paths; keep the final diagnosis with the frontier model.
- 62 
- 63 ## Guardrails
- 64 
- 65 - Do not delegate the immediate blocker if your next step depends on it.
- 66 - Do not ask multiple agents to edit the same files at the same time.
- 67 - Do not trust subagent conclusions blindly when the risk is high; inspect the
- 68   important evidence yourself.
- 69 - Do not claim universal savings. The pattern works best when exploration and
- 70   implementation, testing, or research can be parallelized.
- 71 
- 72 ## Default Framing
- 73 
- 74 "I will use the frontier model as the orchestrator and reviewer, and use
- 75 cheaper subagents for token-heavy research, coding, or testing so the expensive
- 76 tokens go to judgment, synthesis, and final quality."
+ 9 The user asked for /efficient-frontier, an alias of /efficient-orchestrator. Follow the efficient-orchestrator skill exactly; install it alongside this alias.

Compare on GitHub

skills/efficient-orchestrator/SKILL.md

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → f1e12a1b75f3

No earlier version was observed for comparison.

Compare on GitHub

skills/factory/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-babysit-pr/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-collect/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-human-digest/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-lookback/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-recover/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-review-prs/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-ship/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/factory-watchdog/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/plan-arbiter/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/plow-ahead/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/quick-recap/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/read-the-damn-docs/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/rewind/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/stay-within-limits/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/turn-into-app/SKILL.md

Week of Oct 26, 2026 · +34 / −29 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

- 24 - Local scaffolding needs Node.js 22.22.0 or newer (including npm/npx) and
- 25   pnpm.
+ 24 - Local scaffolding needs Node.js 22.22.0 or newer (with npm/npx) and pnpm.
- 28   that checkout with steps 1-6. Never call `start-workspace-app-creation` or
- 29   `create_workspace_app` on this path.
+ 27   that checkout with steps 1-6, unless the route is Builder hosting (below).
+ 28   Never call `create_workspace_app`.
+ 38 
+ 39 ### Ask where the app runs
+ 40 
+ 41 On an interactive local host, after the plan (step 2) and before installing,
+ 42 ask once: **On your machine** (some setup) or **Hosted with Builder** (a free
+ 43 Builder account). Skip it on browser hosts or when the route is set (by the
+ 44 person, a resumed run, Local File Mode, an existing workspace, or headless);
+ 45 [where it runs](references/where-it-runs.md) says which wins.
- 87 Ask once, with your recommended interpretation, only when:
+ 94 Beyond the hosting question, ask once, with your recommendation, only when:
- 115 - [ ] 2 Design decided; brief, App design, and time estimate shown
+ 122 - [ ] 2 Design decided; brief, App design, estimate shown; where it runs asked
- 125 from its git status, brief, schema, and actions. Open your first message with
- 126 what exists, the next stage, and the estimated active work, for example:
- 127 "Found a half-built app in this folder from an earlier run that stopped
- 128 partway. The data model and actions are done, so I'll pick up from there:
- 129 screens next, then checks. About 10 more minutes." Resume at the first
- 130 unverified stage without rerunning scaffolding or replacing completed work.
+ 132 from its git status, brief, schema, and actions. Open with what exists, the
+ 133 next stage, and the active work left ("Found a half-built app. The data model
+ 134 and actions are done; screens and checks are next, about 10 more minutes.").
+ 135 Resume at the first unverified stage without rerunning scaffolding or
+ 136 replacing completed work.
- 167 repeated title, or stat strip over visible data. Before the first scaffold
- 168 command, show the brief with its App design in your message (a saved file is
- 169 not shown), then continue without waiting.
+ 173 repeated title, or stat strip over visible data. Show the brief with its App
+ 174 design in your message (a saved file is not shown), then ask where the app runs
+ 175 and wait, unless that question is skipped.
- 272 - Installed design skills are optional; the review loop limits them.
- 273 
- 274 Screenshots stay in the app's `.tmp/ui-review/out/`.
+ 278 - Screenshots stay in the app's `.tmp/ui-review/out/`. Installed design skills
+ 279   are optional; the review loop limits them.
- 291 which. Label evidence separately:
- 292 locally running, locally verified, build-ready, deployed, and live-verified
- 293 are different states.
+ 296 which. Label evidence separately: locally running, locally verified,
+ 297 build-ready, deployed, and live-verified are different states.
- 297 Keep it demo-short. The first line is the verdict ("Built X in `<dir>`;
- 298 locally verified, not deployed"). Then: the app directory, local URL, and
- 299 server state (PID and stop command); the archetype, the direction, and what
- 300 each control does (agent handoff or local action); absolute paths of the final
- 301 screenshots, rubric scores, and open criteria; the verification, with evidence
- 302 labels, and what the review click left behind; a deployment URL only if it is
- 303 real, and one precise pending step if any; what changed from the brief
- 304 (assumptions, gaps in the source, choices where it was silent).
+ 301 For a local build, keep it demo-short. The first line is the verdict ("Built X
+ 302 in `<dir>`; locally verified, not deployed"). Then: the app directory, local
+ 303 URL, and server state (PID and stop command); the archetype, the direction, and
+ 304 what each control does (agent handoff or local action); absolute paths of the
+ 305 final screenshots, rubric scores, and open criteria; the verification, with
+ 306 evidence labels, and what the review click left behind; a deployment URL only if
+ 307 it is real, and one precise pending step if any; what changed from the brief
+ 308 (assumptions, gaps in the source, choices where it was silent); how to change
+ 309 the app (ask this agent, in this folder).

Compare on GitHub

skills/visual-edit/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/visual-plan/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/visual-recap/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

skills/webmcp/SKILL.md

Week of Oct 26, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

49528d601d43 → f1e12a1b75f3

Compare on GitHub

View on GitHubAdd to my skills

Verified commit f1e12a1b75f342e0b49ebaf182b05168f4fa7b96

git clone --filter=blob:none --no-checkout https://github.com/BuilderIO/skills.git .hotskills/builderio-skills && git -C .hotskills/builderio-skills checkout --detach f1e12a1b75f342e0b49ebaf182b05168f4fa7b96 # Copy each selected skill directory from .hotskills/builderio-skills into .claude/skills/<skill-name>/
4.6kGitHub stars
23stars in 7 days
25skills in this repository
11commits in 30 days

Heat

5.9
  • Star growth, last 7 daysAhead of 71% of tracked repositories
  • Star growth, last 24 hoursAhead of 84% of tracked repositories
  • Rank movement, last 7 daysAhead of 39% of tracked repositories
  • Commits, last 30 daysAhead of 78% 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

Aug 2, 2026Nov 6, 20264,649

Individual skills in this repository

How it works

agent-watchdog

When your agent uses it

Use when asked to watch, babysit, audit, review, compare, or fix another agent's work from a Codex session ID, Claude Code session/transcript, chat/thread link, PR, branch, log, or pasted run summary. Monitor until the other agent is done or blocked, reconstruct what the user asked, independently investigate the same problem to form your own hypotheses and approach, inspect what the agent actually changed and verified, compare the two investigations, report gaps and add-on directions, and optionally make scoped fixes when the user authorizes repair.

What it loads into context

Frontmatter

119 tokens

Always available to the agent

SKILL.md

1,671 tokens

When the skill is activated

Reference documents

103,706 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Choose The Mode
  2. Resolve The Target
  3. Reconstruct The Contract
  4. Investigate Independently
  5. Relay Sparingly
  6. Audit The Evidence
  7. Fix Narrowly
  8. Report

Files

  • ../../.claude-plugin/marketplace.jsonasset1,093 bytes
  • ../../.claude-plugin/plugin.jsonasset672 bytes
  • ../../.mcp.jsonasset138 bytes
  • README.mddoc1,378 bytes
  • SKILL.mdcore7,821 bytes
  • SKILL.mdcore7,821 bytes
  • SKILL.mdcore7,821 bytes
  • SKILL.mdcore7,821 bytes
  • agents/openai.yamlasset210 bytes
  • ../an/SKILL.mddoc4,692 bytes
  • ../efficient-fable/README.mddoc137 bytes
  • ../efficient-fable/SKILL.mddoc367 bytes
  • ../efficient-frontier/README.mddoc140 bytes
  • ../efficient-frontier/SKILL.mddoc379 bytes
  • ../efficient-orchestrator/README.mddoc1,304 bytes
  • ../efficient-orchestrator/SKILL.mddoc2,256 bytes
  • ../efficient-orchestrator/assets/fable-orchestrator-dark.pngasset111,350 bytes
  • ../efficient-orchestrator/assets/fable-orchestrator.excalidrawasset19,562 bytes
  • ../efficient-orchestrator/assets/fable-orchestrator.pngasset111,227 bytes
  • ../factory-babysit-pr/README.mddoc827 bytes
  • ../factory-babysit-pr/SKILL.mddoc3,610 bytes
  • ../factory-collect/SKILL.mddoc3,225 bytes
  • ../factory-human-digest/SKILL.mddoc3,924 bytes
  • ../factory-lookback/SKILL.mddoc4,073 bytes
  • ../factory-recover/SKILL.mddoc1,685 bytes
  • ../factory-review-prs/SKILL.mddoc2,445 bytes
  • ../factory-ship/SKILL.mddoc2,153 bytes
  • ../factory-watchdog/SKILL.mddoc1,843 bytes
  • ../factory/SKILL.mddoc3,998 bytes
  • ../plan-arbiter/README.mddoc1,353 bytes
  • ../plan-arbiter/SKILL.mddoc4,085 bytes
  • ../plan-arbiter/agents/openai.yamlasset224 bytes
  • ../plow-ahead/README.mddoc1,387 bytes
  • ../plow-ahead/SKILL.mddoc3,889 bytes
  • ../plow-ahead/agents/openai.yamlasset260 bytes
  • ../quick-recap/README.mddoc1,442 bytes
  • ../quick-recap/SKILL.mddoc1,420 bytes
  • ../read-the-damn-docs/README.mddoc2,291 bytes
  • ../read-the-damn-docs/SKILL.mddoc7,029 bytes
  • ../read-the-damn-docs/agents/openai.yamlasset215 bytes
  • ../rewind/README.mddoc3,859 bytes
  • ../rewind/SKILL.mddoc4,059 bytes
  • ../stay-within-limits/README.mddoc2,423 bytes
  • ../stay-within-limits/SKILL.mddoc3,427 bytes
  • ../turn-into-app/README.mddoc4,802 bytes
  • ../turn-into-app/SKILL.mddoc16,932 bytes
  • ../turn-into-app/agents/openai.yamlasset383 bytes
  • ../turn-into-app/references/attachments.mddoc1,503 bytes
  • ../turn-into-app/references/fresh-project.mddoc4,733 bytes
  • ../turn-into-app/references/local-run-and-deploy.mddoc7,649 bytes
  • ../turn-into-app/references/review-loop.mddoc19,749 bytes
  • ../turn-into-app/references/source-brief.mddoc6,588 bytes
  • ../turn-into-app/references/spreadsheet-source.mddoc14,401 bytes
  • ../turn-into-app/references/ui-archetypes.mddoc19,180 bytes
  • ../turn-into-app/references/ui-direction.mddoc19,431 bytes
  • ../turn-into-app/references/ui-palettes.mddoc23,241 bytes
  • ../turn-into-app/references/where-it-runs.mddoc4,905 bytes
  • ../visual-edit/README.mddoc4,159 bytes
  • ../visual-edit/SKILL.mddoc30,968 bytes
  • ../visual-plan/README.mddoc4,005 bytes
  • ../visual-plan/SKILL.mddoc32,771 bytes
  • ../visual-plan/references/canvas.mddoc9,388 bytes
  • ../visual-plan/references/connection.mddoc2,980 bytes
  • ../visual-plan/references/document-quality.mddoc12,638 bytes
  • ../visual-plan/references/exemplar.mddoc4,028 bytes
  • ../visual-plan/references/local-files.mddoc6,509 bytes
  • ../visual-plan/references/wireframe.mddoc19,471 bytes
  • ../visual-recap/README.mddoc4,234 bytes
  • ../visual-recap/SKILL.mddoc33,616 bytes
  • ../visual-recap/references/connection.mddoc2,980 bytes
  • ../visual-recap/references/local-files.mddoc6,509 bytes
  • ../visual-recap/references/wireframe.mddoc19,471 bytes
  • ../webmcp/README.mddoc3,378 bytes
  • ../webmcp/SKILL.mddoc12,966 bytes
  • ../webmcp/agents/openai.yamlasset1,228 bytes
Read SKILL.md

# Agent Watchdog

Watch another agent's work like a reviewer with a pager who is also a second
investigator: wait for completion when needed, reconstruct the request, run
your own independent investigation of the same problem, verify the evidence,
and close the gap between what was asked and what actually happened. You are
not just grading their homework — you are a second solver whose findings get
diffed against theirs so nothing is missed.

## Choose The Mode

Infer the mode from the user's wording:

- **Watch only:** monitor a session, PR, branch, CI run, or transcript until it
View the complete source on GitHub

Skills included

  • agent-watchdog
  • an
  • efficient-fable
  • efficient-frontier
  • efficient-orchestrator
  • factory
  • factory-babysit-pr
  • factory-collect
  • factory-human-digest
  • factory-lookback
  • factory-recover
  • factory-review-prs
  • factory-ship
  • factory-watchdog
  • plan-arbiter
  • plow-ahead
  • quick-recap
  • read-the-damn-docs
  • rewind
  • stay-within-limits
  • turn-into-app
  • visual-edit
  • visual-plan
  • visual-recap
  • webmcp

Repository facts

License
MIT
Created
Jun 10, 2026
Last pushed
Nov 2, 2026
Forks
236
Open issues
3
Project activity
Active

Rank over the last 12 weeks

Now #292 · best #230 · down 37 in 30 days

  1. #281 on Oct 29
  2. #283 on Oct 30
  3. #284 from Oct 31 to Nov 2
  4. #287 on Nov 3
  5. #289 on Nov 4
  6. #292 since Nov 5
67 earlier periods
  1. #225 on Aug 2
  2. #228 from Aug 7 to Aug 8
  3. #225 on Aug 9
  4. #232 on Aug 10
  5. #230 on Aug 11
  6. #229 on Aug 12
  7. #231 from Aug 13 to Aug 14
  8. #230 from Aug 15 to Aug 17
  9. #238 on Aug 18
  10. #235 from Aug 19 to Aug 20
  11. #236 on Aug 21
  12. #234 on Aug 22
  13. #238 on Aug 23
  14. #239 on Aug 24
  15. #238 from Aug 25 to Aug 26
  16. #239 on Aug 27
  17. #237 from Aug 28 to Aug 29
  18. #236 on Aug 30
  19. #235 on Aug 31
  20. #241 on Sep 1
  21. #235 on Sep 2
  22. #243 on Sep 3
  23. #238 on Sep 4
  24. #241 on Sep 5
  25. #239 on Sep 6
  26. #241 on Sep 7
  27. #242 on Sep 8
  28. #243 on Sep 9
  29. #241 on Sep 10
  30. #242 from Sep 11 to Sep 12
  31. #246 on Sep 13
  32. #245 on Sep 14
  33. #246 on Sep 15
  34. #244 from Sep 16 to Sep 17
  35. #249 on Sep 18
  36. #245 on Sep 19
  37. #250 from Sep 20 to Sep 21
  38. #249 on Sep 22
  39. #250 from Sep 23 to Sep 24
  40. #247 on Sep 25
  41. #249 on Sep 26
  42. #248 on Sep 27
  43. #251 on Sep 28
  44. #252 on Sep 29
  45. #249 on Sep 30
  46. #251 on Oct 1
  47. #252 on Oct 2
  48. #249 on Oct 3
  49. #246 on Oct 4
  50. #250 on Oct 5
  51. #252 on Oct 6
  52. #255 from Oct 7 to Oct 8
  53. #259 on Oct 9
  54. #256 on Oct 10
  55. #258 on Oct 11
  56. #262 on Oct 12
  57. #263 on Oct 13
  58. #261 on Oct 14
  59. #264 on Oct 15
  60. #261 on Oct 16
  61. #264 from Oct 17 to Oct 18
  62. #265 on Oct 19
  63. #269 from Oct 21 to Oct 22
  64. #268 on Oct 23
  65. #269 on Oct 24
  66. #271 on Oct 25
  67. #276 on Oct 28

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 2, 2026

Changed on Oct 27, 2026

No longer found: Access to secrets

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 97 notes
  • skills/agent-watchdog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/agent-watchdog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/agent-watchdog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-collect/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/factory-human-digest/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 330 signals by file

.claude-plugin/marketplace.json

  • Network accessline 3
    "owner": {     "name": "Builder.io",     "url": "https://www.builder.io"
  • Network accessline 4
    "name": "Builder.io",     "url": "https://www.builder.io"   },
  • Network accessline 5
    "url": "https://www.builder.io"
  • Network accessline 14
    "author": {         "name": "Builder.io",         "url": "https://www.builder.io"
  • Network accessline 15
    "name": "Builder.io",         "url": "https://www.builder.io"       },
  • Network accessline 16
    "url": "https://www.builder.io"
  • Network accessline 17
    },       "homepage": "https://github.com/BuilderIO/skills",       "repository": "https://github.com/BuilderIO/skills",
  • Network accessline 18
    "homepage": "https://github.com/BuilderIO/skills",
  • Network accessline 19
    "repository": "https://github.com/BuilderIO/skills",

.claude-plugin/plugin.json

  • Network accessline 5
    "author": {     "name": "Builder.io",     "url": "https://www.builder.io"
  • Network accessline 6
    "name": "Builder.io",     "url": "https://www.builder.io"   },
  • Network accessline 7
    "url": "https://www.builder.io"
  • Network accessline 8
    },   "homepage": "https://github.com/BuilderIO/skills",   "repository": "https://github.com/BuilderIO/skills",
  • Network accessline 9
    "homepage": "https://github.com/BuilderIO/skills",
  • Network accessline 10
    "repository": "https://github.com/BuilderIO/skills",

.mcp.json

  • Network accessline 3
    "agent-native-dispatch": {       "type": "http",       "url": "https://dispatch.agent-native.com/mcp"
  • Network accessline 4
    "type": "http",       "url": "https://dispatch.agent-native.com/mcp"     }
  • Network accessline 5
    "url": "https://dispatch.agent-native.com/mcp"

skills/an/SKILL.md

  • Network accessline 82
    The generic Agent-Native plugin registers Dispatch at `https://dispatch.agent-native.com/mcp` for hosts that support plugin MCP
  • Network accessline 82
    The generic Agent-Native plugin registers Dispatch at `https://dispatch.agent-native.com/mcp` for hosts that support plugin MCP
  • Network accessline 83
    `https://dispatch.agent-native.com/mcp` for hosts that support plugin MCP
  • Network accessline 83
    `https://dispatch.agent-native.com/mcp` for hosts that support plugin MCP

skills/efficient-orchestrator/assets/fable-orchestrator.excalidraw

  • Network accessline 2
    "type": "excalidraw",   "version": 2,   "source": "https://excalidraw.com",
  • Network accessline 2
    "type": "excalidraw",   "version": 2,   "source": "https://excalidraw.com",
  • Network accessline 3
    "version": 2,   "source": "https://excalidraw.com",   "elements": [
  • Network accessline 3
    "version": 2,   "source": "https://excalidraw.com",   "elements": [
  • Network accessline 4
    "source": "https://excalidraw.com",
  • Network accessline 4
    "source": "https://excalidraw.com",

skills/factory-babysit-pr/SKILL.md

  • Network accessline 19
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-babysit-pr` entry, the target repository's instructions, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configurati
  • Network accessline 19
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-babysit-pr` entry, the target repository's instructions, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configurati
  • Network accessline 20
    `skill_prompts.factory-babysit-pr` entry, the target repository's instructions, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md#workflowspr-babysitting). A PR number, lab
  • Network accessline 20
    `skill_prompts.factory-babysit-pr` entry, the target repository's instructions, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md#workflowspr-babysitting). A PR number, lab
  • Network accessline 21
    and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md#workflowspr-babysitting).
  • Network accessline 21
    and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md#workflowspr-babysitting).

skills/factory-recover/SKILL.md

  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-recover` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Recover only configured projects
  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-recover` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Recover only configured projects
  • Network accessline 15
    `skill_prompts.factory-recover` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Recover only configured projects and workflow types; do not restart every stopped
  • Network accessline 15
    `skill_prompts.factory-recover` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Recover only configured projects and workflow types; do not restart every stopped

skills/factory-review-prs/SKILL.md

  • Network accessline 14
    This skill reviews a filtered queue. For one long-lived, explicitly authorized PR, use `factory-babysit-pr`. Read the filters and independent action policies from `.agent-factory/config.yaml`; see the [configuration reference](https://githu
  • Network accessline 14
    This skill reviews a filtered queue. For one long-lived, explicitly authorized PR, use `factory-babysit-pr`. Read the filters and independent action policies from `.agent-factory/config.yaml`; see the [configuration reference](https://githu
  • Network accessline 15
    PR, use `factory-babysit-pr`. Read the filters and independent action policies from `.agent-factory/config.yaml`; see the [configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Apply the opti
  • Network accessline 15
    PR, use `factory-babysit-pr`. Read the filters and independent action policies from `.agent-factory/config.yaml`; see the [configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Apply the opti
  • Network accessline 16
    from `.agent-factory/config.yaml`; see the [configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md).
  • Network accessline 16
    from `.agent-factory/config.yaml`; see the [configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md).

skills/factory-ship/SKILL.md

  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-ship` entry, and the repository's instructions before publishing. See the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/co
  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-ship` entry, and the repository's instructions before publishing. See the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/co
  • Network accessline 16
    publishing. See the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). The prompt adds project guidance; editing code does not by itself authorize a PR update, approval, merge, or
  • Network accessline 16
    publishing. See the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). The prompt adds project guidance; editing code does not by itself authorize a PR update, approval, merge, or

skills/factory-watchdog/SKILL.md

  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-watchdog` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Monitor only configured projects
  • Network accessline 14
    Read `.agent-factory/config.yaml`, the optional `skill_prompts.factory-watchdog` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Monitor only configured projects
  • Network accessline 15
    `skill_prompts.factory-watchdog` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Monitor only configured projects and authorized delivery handoffs. The prompt ca
  • Network accessline 15
    `skill_prompts.factory-watchdog` entry, and the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md). Monitor only configured projects and authorized delivery handoffs. The prompt ca

skills/factory/SKILL.md

  • Network accessline 23
    Use the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md) for field meanings, examples, and host limitations.
  • Network accessline 23
    Use the [Factory configuration reference](https://github.com/BuilderIO/skills/blob/main/docs/factory/configuration.md) for field meanings, examples, and host limitations.

skills/rewind/README.md

  • Network accessline 10
    - macOS with the current signed [Clips Desktop](https://clips.agent-native.com/download)
  • Network accessline 10
    - macOS with the current signed [Clips Desktop](https://clips.agent-native.com/download)
  • Network accessline 35
    After permission, a capable agent can open the official install flow. If the agent cannot control native UI, it should give you the direct [Clips Desktop download link](https://clips.agent-native.com/download) and wait
  • Network accessline 35
    After permission, a capable agent can open the official install flow. If the agent cannot control native UI, it should give you the direct [Clips Desktop download link](https://clips.agent-native.com/download) and wait
  • Network accessline 36
    agent cannot control native UI, it should give you the direct [Clips Desktop download link](https://clips.agent-native.com/download) and wait for you to finish. Then:
  • Network accessline 36
    agent cannot control native UI, it should give you the direct [Clips Desktop download link](https://clips.agent-native.com/download) and wait for you to finish. Then:
  • Network accessline 37
    [Clips Desktop download link](https://clips.agent-native.com/download) and wait
  • Network accessline 37
    [Clips Desktop download link](https://clips.agent-native.com/download) and wait
  • Network accessline 78
    - If Clips Desktop is missing, ask the agent to open the official   [download page](https://clips.agent-native.com/download), or open it yourself.
  • Network accessline 78
    - If Clips Desktop is missing, ask the agent to open the official   [download page](https://clips.agent-native.com/download), or open it yourself.
  • Network accessline 79
    [download page](https://clips.agent-native.com/download), or open it yourself.
  • Network accessline 79
    [download page](https://clips.agent-native.com/download), or open it yourself.
  • Network accessline 88
    See the [Clips documentation](https://www.agent-native.com/docs/template-clips)
  • Network accessline 88
    See the [Clips documentation](https://www.agent-native.com/docs/template-clips)
  • Network accessline 89
    and [Agent Native source](https://github.com/BuilderIO/agent-native) for the
  • Network accessline 89
    and [Agent Native source](https://github.com/BuilderIO/agent-native) for the

skills/rewind/SKILL.md

  • Network accessline 29
    through setup?” 3. Only after affirmative permission, open    `https://clips.agent-native.com/download` in the user's browser. Complete
  • Network accessline 29
    through setup?” 3. Only after affirmative permission, open    `https://clips.agent-native.com/download` in the user's browser. Complete
  • Network accessline 30
    3. Only after affirmative permission, open    `https://clips.agent-native.com/download` in the user's browser. Complete    native installation steps only when the current host can safely do so and
  • Network accessline 30
    3. Only after affirmative permission, open    `https://clips.agent-native.com/download` in the user's browser. Complete    native installation steps only when the current host can safely do so and
  • Network accessline 31
    `https://clips.agent-native.com/download` in the user's browser. Complete
  • Network accessline 31
    `https://clips.agent-native.com/download` in the user's browser. Complete

skills/turn-into-app/README.md

  • Network accessline 3
    Give a workflow you have already proven the face of an app and the brain of an agent. `/turn-into-app` turns a thread, a skill, a spreadsheet, or a Claude or ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a
  • Network accessline 3
    Give a workflow you have already proven the face of an app and the brain of an agent. `/turn-into-app` turns a thread, a skill, a spreadsheet, or a Claude or ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a
  • Network accessline 4
    agent. `/turn-into-app` turns a thread, a skill, a spreadsheet, or a Claude or ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a populated, domain-shaped interface (a week grid, a board, a live workbench, a
  • Network accessline 4
    agent. `/turn-into-app` turns a thread, a skill, a spreadsheet, or a Claude or ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a populated, domain-shaped interface (a week grid, a board, a live workbench, a
  • Network accessline 5
    ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a
  • Network accessline 5
    ChatGPT project into a visual [Agent-Native](https://agent-native.com) app: a

skills/turn-into-app/references/attachments.md

  • Network accessline 39
    {   type: "url",   value: "https://..."
  • Network accessline 39
    {   type: "url",   value: "https://..."
  • Network accessline 40
    type: "url",   value: "https://..." }
  • Network accessline 40
    type: "url",   value: "https://..." }
  • Network accessline 41
    value: "https://..."
  • Network accessline 41
    value: "https://..."

skills/turn-into-app/references/fresh-project.md

  • Network accessline 69
    instructions in the Project instructions or knowledge files. - Add and authenticate the Agent-Native Dispatch MCP connector   (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the
  • Network accessline 69
    instructions in the Project instructions or knowledge files. - Add and authenticate the Agent-Native Dispatch MCP connector   (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the
  • Network accessline 70
    - Add and authenticate the Agent-Native Dispatch MCP connector   (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the   workspace's Builder project through the Builder Projects API, so no separate
  • Network accessline 70
    - Add and authenticate the Agent-Native Dispatch MCP connector   (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the   workspace's Builder project through the Builder Projects API, so no separate
  • Network accessline 71
    (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the
  • Network accessline 71
    (`https://dispatch.agent-native.com/mcp`, OAuth). It reuses or provisions the

skills/turn-into-app/references/local-run-and-deploy.md

  • Network accessline 52
    defaults only. Precedence, supported modes, and the line between committed config and deployment secrets: https://www.agent-native.com/docs/agent-native-config
  • Network accessline 52
    defaults only. Precedence, supported modes, and the line between committed config and deployment secrets: https://www.agent-native.com/docs/agent-native-config
  • Network accessline 54
    https://www.agent-native.com/docs/agent-native-config
  • Network accessline 54
    https://www.agent-native.com/docs/agent-native-config
  • Access to secretsline 71
    [Credential-related source omitted]
  • Access to secretsline 71
    [Credential-related source omitted]
  • Access to secretsline 72
    [Credential-related source omitted]
  • Access to secretsline 72
    [Credential-related source omitted]
  • Network accessline 85
    mkdir -p .tmp && (nohup pnpm exec agent-native dev --port <port> > .tmp/dev.log 2>&1 & echo $! > .tmp/dev.pid) wait200() { while t=$((end-SECONDS)); [ $t -gt 0 ]; do [ "$(curl -sL --connect-timeout 2 --max-time $((t<10?t:10)) -o /dev/null -
  • Network accessline 85
    mkdir -p .tmp && (nohup pnpm exec agent-native dev --port <port> > .tmp/dev.log 2>&1 & echo $! > .tmp/dev.pid) wait200() { while t=$((end-SECONDS)); [ $t -gt 0 ]; do [ "$(curl -sL --connect-timeout 2 --max-time $((t<10?t:10)) -o /dev/null -
  • Network accessline 86
    wait200() { while t=$((end-SECONDS)); [ $t -gt 0 ]; do [ "$(curl -sL --connect-timeout 2 --max-time $((t<10?t:10)) -o /dev/null -w '%{http_code}' "$1")" = 200 ] && return 0; sleep 1; done; echo "no 200 from $1 before the three-minute deadli
  • Network accessline 86
    wait200() { while t=$((end-SECONDS)); [ $t -gt 0 ]; do [ "$(curl -sL --connect-timeout 2 --max-time $((t<10?t:10)) -o /dev/null -w '%{http_code}' "$1")" = 200 ] && return 0; sleep 1; done; echo "no 200 from $1 before the three-minute deadli
  • Network accessline 87
    end=$((SECONDS+180)); wait200 http://localhost:<port>/ && wait200 http://localhost:<port>/<route>
  • Network accessline 87
    end=$((SECONDS+180)); wait200 http://localhost:<port>/ && wait200 http://localhost:<port>/<route>
  • Network accessline 92
    assignment, and each request is capped by the time left; run that last line again after every restart for a fresh clock. A nonzero exit means the server or the route never answered: read `.tmp/dev.log` and fix that before any screenshot. Th
  • Network accessline 92
    assignment, and each request is capped by the time left; run that last line again after every restart for a fresh clock. A nonzero exit means the server or the route never answered: read `.tmp/dev.log` and fix that before any screenshot. Th
  • Network accessline 93
    again after every restart for a fresh clock. A nonzero exit means the server or the route never answered: read `.tmp/dev.log` and fix that before any screenshot. The log prints `Local: http://localhost:<port>/` before the server can answer;
  • Network accessline 93
    again after every restart for a fresh clock. A nonzero exit means the server or the route never answered: read `.tmp/dev.log` and fix that before any screenshot. The log prints `Local: http://localhost:<port>/` before the server can answer;
  • Network accessline 94
    read `.tmp/dev.log` and fix that before any screenshot. The log prints `Local: http://localhost:<port>/` before the
  • Network accessline 94
    read `.tmp/dev.log` and fix that before any screenshot. The log prints `Local: http://localhost:<port>/` before the
  • Network accessline 137
    `pnpm build`, then the provider's own deploy command. Local PGlite storage is   not production storage. Guide:   https://www.agent-native.com/docs/deploy-an-app
  • Network accessline 137
    `pnpm build`, then the provider's own deploy command. Local PGlite storage is   not production storage. Guide:   https://www.agent-native.com/docs/deploy-an-app
  • Network accessline 138
    not production storage. Guide:   https://www.agent-native.com/docs/deploy-an-app - **App inside a workspace:** deploy the workspace from its root, which needs
  • Network accessline 138
    not production storage. Guide:   https://www.agent-native.com/docs/deploy-an-app - **App inside a workspace:** deploy the workspace from its root, which needs
  • Network accessline 139
    https://www.agent-native.com/docs/deploy-an-app
  • Network accessline 139
    https://www.agent-native.com/docs/deploy-an-app
  • Network accessline 148
    For Vercel, run `npx @agent-native/core@latest deploy --preset vercel`, then   `vercel deploy --prebuilt`. Guide:   https://www.agent-native.com/docs/workspace-deployment
  • Network accessline 148
    For Vercel, run `npx @agent-native/core@latest deploy --preset vercel`, then   `vercel deploy --prebuilt`. Guide:   https://www.agent-native.com/docs/workspace-deployment
  • Network accessline 150
    https://www.agent-native.com/docs/workspace-deployment
  • Network accessline 150
    https://www.agent-native.com/docs/workspace-deployment

skills/turn-into-app/references/review-loop.md

  • Network accessline 39
    node .tmp/ui-review/shoot.mjs http://localhost:<port>/ p1 'button:has-text("<agent verb>")' /<route>
  • Network accessline 39
    node .tmp/ui-review/shoot.mjs http://localhost:<port>/ p1 'button:has-text("<agent verb>")' /<route>

skills/turn-into-app/references/spreadsheet-source.md

  • Network accessline 50
    host, or, in Dispatch or Analytics, the `google_drive` provider through    `provider-api-catalog`, `provider-api-docs`, and `provider-api-request`    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the
  • Network accessline 50
    host, or, in Dispatch or Analytics, the `google_drive` provider through    `provider-api-catalog`, `provider-api-docs`, and `provider-api-request`    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the
  • Network accessline 51
    `provider-api-catalog`, `provider-api-docs`, and `provider-api-request`    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the    provider's base is Drive v3). The chat scaffold's `provider-api-request` is
  • Network accessline 51
    `provider-api-catalog`, `provider-api-docs`, and `provider-api-request`    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the    provider's base is Drive v3). The chat scaffold's `provider-api-request` is
  • Network accessline 52
    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the
  • Network accessline 52
    with a full `https://sheets.googleapis.com/v4/spreadsheets/<id>` URL (the

skills/turn-into-app/references/where-it-runs.md

  • Network accessline 78
    1. Connect Dispatch. If it is not configured, run    `npx @agent-native/core@latest connect https://dispatch.agent-native.com`
  • Network accessline 78
    1. Connect Dispatch. If it is not configured, run    `npx @agent-native/core@latest connect https://dispatch.agent-native.com`
  • Network accessline 79
    `npx @agent-native/core@latest connect https://dispatch.agent-native.com`
  • Network accessline 79
    `npx @agent-native/core@latest connect https://dispatch.agent-native.com`
  • Network accessline 80
    with `--client` set to this host. Without Node.js 22.22+, don't make the    person install it just to pick Builder: send them to    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each
  • Network accessline 80
    with `--client` set to this host. Without Node.js 22.22+, don't make the    person install it just to pick Builder: send them to    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each
  • Network accessline 81
    person install it just to pick Builder: send them to    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each    host. If it is configured but signed out, sign in from the host's MCP
  • Network accessline 81
    person install it just to pick Builder: send them to    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each    host. If it is configured but signed out, sign in from the host's MCP
  • Network accessline 82
    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each
  • Network accessline 82
    `https://dispatch.agent-native.com/mcp/connect`, which has steps for each

skills/visual-edit/README.md

  • Network accessline 60
    "title": "Checkout flow review",   "routes": [     { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },
  • Network accessline 60
    "title": "Checkout flow review",   "routes": [     { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },
  • Network accessline 61
    "routes": [     { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },     { "url": "http://localhost:5173/checkout?step=payment", "title": "2. Payment" },
  • Network accessline 61
    "routes": [     { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },     { "url": "http://localhost:5173/checkout?step=payment", "title": "2. Payment" },
  • Network accessline 62
    { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },
  • Network accessline 62
    { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },
  • Network accessline 63
    { "url": "http://localhost:5173/checkout?step=payment", "title": "2. Payment" },
  • Network accessline 63
    { "url": "http://localhost:5173/checkout?step=payment", "title": "2. Payment" },
  • Network accessline 64
    { "url": "http://localhost:5173/checkout?step=done", "title": "3. Confirmation" }
  • Network accessline 64
    { "url": "http://localhost:5173/checkout?step=done", "title": "3. Confirmation" }
  • Network accessline 95
    For the complete walkthrough and sharing details, see the [Visual Edit guide in Agent-Native](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md).
  • Network accessline 95
    For the complete walkthrough and sharing details, see the [Visual Edit guide in Agent-Native](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md).
  • Network accessline 96
    in Agent-Native](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md).
  • Network accessline 96
    in Agent-Native](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md).

skills/visual-edit/SKILL.md

  • Network accessline 18
    The editor is hosted at `https://design.agent-native.com`. Never start local
  • Network accessline 18
    The editor is hosted at `https://design.agent-native.com`. Never start local
  • Network accessline 21
    See the [Visual Edit guide](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md)
  • Network accessline 21
    See the [Visual Edit guide](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md)
  • Network accessline 57
    - Inline-browser hosts should open `https://design.agent-native.com/visual-edit`
  • Network accessline 57
    - Inline-browser hosts should open `https://design.agent-native.com/visual-edit`
  • Network accessline 73
    JavaScript evaluator. Prefer the host's inline browser; use external Chrome only when requested or unavailable. Open `https://design.agent-native.com/visual-edit`, keep it visible while tools
  • Network accessline 73
    JavaScript evaluator. Prefer the host's inline browser; use external Chrome only when requested or unavailable. Open `https://design.agent-native.com/visual-edit`, keep it visible while tools
  • Network accessline 74
    only when requested or unavailable. Open `https://design.agent-native.com/visual-edit`, keep it visible while tools register, and read the title before work. Never enter, copy, or request
  • Network accessline 74
    only when requested or unavailable. Open `https://design.agent-native.com/visual-edit`, keep it visible while tools register, and read the title before work. Never enter, copy, or request
  • Network accessline 75
    `https://design.agent-native.com/visual-edit`, keep it visible while tools
  • Network accessline 75
    `https://design.agent-native.com/visual-edit`, keep it visible while tools
  • Network accessline 78
    other pages that block tools behind sign-in still require a signed-in tab. If the browser exposes CDP permissions, grant `local-network-access` to `https://design.agent-native.com` before calling page tools; otherwise use the
  • Network accessline 78
    other pages that block tools behind sign-in still require a signed-in tab. If the browser exposes CDP permissions, grant `local-network-access` to `https://design.agent-native.com` before calling page tools; otherwise use the
  • Network accessline 79
    If the browser exposes CDP permissions, grant `local-network-access` to `https://design.agent-native.com` before calling page tools; otherwise use the page's Connect button and let the browser's permission prompt complete.
  • Network accessline 79
    If the browser exposes CDP permissions, grant `local-network-access` to `https://design.agent-native.com` before calling page tools; otherwise use the page's Connect button and let the browser's permission prompt complete.
  • Network accessline 80
    `https://design.agent-native.com` before calling page tools; otherwise use the
  • Network accessline 80
    `https://design.agent-native.com` before calling page tools; otherwise use the
  • Network accessline 99
    } const result = await an.call("open-visual-edit", {   devServerUrl: "http://localhost:5173",
  • Network accessline 99
    } const result = await an.call("open-visual-edit", {   devServerUrl: "http://localhost:5173",
  • Network accessline 100
    const result = await an.call("open-visual-edit", {   devServerUrl: "http://localhost:5173",   paths: ["/"],
  • Network accessline 100
    const result = await an.call("open-visual-edit", {   devServerUrl: "http://localhost:5173",   paths: ["/"],
  • Network accessline 101
    devServerUrl: "http://localhost:5173",
  • Network accessline 101
    devServerUrl: "http://localhost:5173",
  • Network accessline 175
    - Browser hosts can use page-local `get-visual-edit-prompt`. - The `open-visual-edit` action is owned by Design. From another app, use the   hosted MCP server at `https://design.agent-native.com/mcp` or the page's
  • Network accessline 175
    - Browser hosts can use page-local `get-visual-edit-prompt`. - The `open-visual-edit` action is owned by Design. From another app, use the   hosted MCP server at `https://design.agent-native.com/mcp` or the page's
  • Network accessline 176
    - The `open-visual-edit` action is owned by Design. From another app, use the   hosted MCP server at `https://design.agent-native.com/mcp` or the page's   WebMCP helper, not `pnpm action` in the target app. The page path works
  • Network accessline 176
    - The `open-visual-edit` action is owned by Design. From another app, use the   hosted MCP server at `https://design.agent-native.com/mcp` or the page's   WebMCP helper, not `pnpm action` in the target app. The page path works
  • Network accessline 177
    hosted MCP server at `https://design.agent-native.com/mcp` or the page's
  • Network accessline 177
    hosted MCP server at `https://design.agent-native.com/mcp` or the page's
  • Network accessline 198
    path/query for another state; preserve the order of named or numbered flows.   Shorthand like `localhost:1234/onboarding/1` means   `http://localhost:1234/onboarding/1`.
  • Network accessline 198
    path/query for another state; preserve the order of named or numbered flows.   Shorthand like `localhost:1234/onboarding/1` means   `http://localhost:1234/onboarding/1`.
  • Network accessline 200
    `http://localhost:1234/onboarding/1`.
  • Network accessline 200
    `http://localhost:1234/onboarding/1`.
  • Network accessline 279
    [Credential-related source omitted]
  • Network accessline 279
    [Credential-related source omitted]
  • Network accessline 280
    [Credential-related source omitted]
  • Network accessline 280
    [Credential-related source omitted]
  • Access to secretsline 280
    [Credential-related source omitted]
  • Access to secretsline 280
    [Credential-related source omitted]
  • Network accessline 291
    npx @agent-native/core@latest design connect --url http://localhost:5173 --root . --json
  • Network accessline 291
    npx @agent-native/core@latest design connect --url http://localhost:5173 --root . --json
  • Network accessline 303
    pnpm dev:cli design connect --url http://localhost:5173 --root templates/<app> --json
  • Network accessline 303
    pnpm dev:cli design connect --url http://localhost:5173 --root templates/<app> --json
  • Network accessline 315
    [Credential-related source omitted]
  • Network accessline 315
    [Credential-related source omitted]
  • Network accessline 318
    (Equivalently, pass `--bridge-token <token>`.) This starts a detached bridge on `http://127.0.0.1:7331`, adopts the server-minted token — so bridge and row
  • Network accessline 318
    (Equivalently, pass `--bridge-token <token>`.) This starts a detached bridge on `http://127.0.0.1:7331`, adopts the server-minted token — so bridge and row
  • Network accessline 319
    `http://127.0.0.1:7331`, adopts the server-minted token — so bridge and row
  • Network accessline 319
    `http://127.0.0.1:7331`, adopts the server-minted token — so bridge and row
  • Network accessline 326
    curl http://127.0.0.1:7331/health
  • Network accessline 326
    curl http://127.0.0.1:7331/health
  • Network accessline 371
    {   "title": "Docs homepage visual edit",   "devServerUrl": "http://localhost:5173",
  • Network accessline 371
    {   "title": "Docs homepage visual edit",   "devServerUrl": "http://localhost:5173",
  • Network accessline 372
    "title": "Docs homepage visual edit",   "devServerUrl": "http://localhost:5173",   "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 372
    "title": "Docs homepage visual edit",   "devServerUrl": "http://localhost:5173",   "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 373
    "devServerUrl": "http://localhost:5173",
  • Network accessline 373
    "devServerUrl": "http://localhost:5173",
  • Network accessline 374
    "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 374
    "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 394
    {   "title": "Tasks responsive visual edit",   "devServerUrl": "http://localhost:5173",
  • Network accessline 394
    {   "title": "Tasks responsive visual edit",   "devServerUrl": "http://localhost:5173",
  • Network accessline 395
    "title": "Tasks responsive visual edit",   "devServerUrl": "http://localhost:5173",   "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 395
    "title": "Tasks responsive visual edit",   "devServerUrl": "http://localhost:5173",   "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 396
    "devServerUrl": "http://localhost:5173",
  • Network accessline 396
    "devServerUrl": "http://localhost:5173",
  • Network accessline 397
    "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 397
    "bridgeUrl": "http://127.0.0.1:7331",
  • Network accessline 431
    "designId": "<existing-design-id>",   "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:5173",
  • Network accessline 431
    "designId": "<existing-design-id>",   "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:5173",
  • Network accessline 432
    "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:5173",   "paths": ["/settings", "/team"],
  • Network accessline 432
    "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:5173",   "paths": ["/settings", "/team"],
  • Network accessline 433
    "devServerUrl": "http://localhost:5173",
  • Network accessline 433
    "devServerUrl": "http://localhost:5173",
  • Network accessline 447
    "designId": "<existing-design-id>",   "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:1234",
  • Network accessline 447
    "designId": "<existing-design-id>",   "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:1234",
  • Network accessline 448
    "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:1234",   "routes": [
  • Network accessline 448
    "connectionId": "<existing-connection-id>",   "devServerUrl": "http://localhost:1234",   "routes": [
  • Network accessline 449
    "devServerUrl": "http://localhost:1234",
  • Network accessline 449
    "devServerUrl": "http://localhost:1234",
  • Network accessline 470
    The fallback still targets `https://design.agent-native.com`; only the app and
  • Network accessline 470
    The fallback still targets `https://design.agent-native.com`; only the app and

skills/visual-plan/README.md

  • Network accessline 19
    Visual plans are MDX, customizable with your own components, and viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). The
  • Network accessline 19
    Visual plans are MDX, customizable with your own components, and viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). The
  • Network accessline 20
    [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). The
  • Network accessline 20
    [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). The
  • Network accessline 22
    MDX locally, starts a localhost bridge, and opens the hosted Plan UI with no sharing. [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 22
    MDX locally, starts a localhost bridge, and opens the hosted Plan UI with no sharing. [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 24
    [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 24
    [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 51
    For teams wiring visual plans into their review flow, the [Plan app documentation](https://www.agent-native.com/docs/template-plan)
  • Network accessline 51
    For teams wiring visual plans into their review flow, the [Plan app documentation](https://www.agent-native.com/docs/template-plan)
  • Network accessline 52
    [Plan app documentation](https://www.agent-native.com/docs/template-plan)
  • Network accessline 52
    [Plan app documentation](https://www.agent-native.com/docs/template-plan)
  • Network accessline 92
    The skill expects the [Plan MCP connector](https://www.agent-native.com/docs/template-plan)
  • Network accessline 92
    The skill expects the [Plan MCP connector](https://www.agent-native.com/docs/template-plan)

skills/visual-plan/SKILL.md

  • Network accessline 509
    separate slash commands. Pass `--no-connect` to register the connector without authenticating, then run `npx @agent-native/core@latest connect https://plan.agent-native.com --client all`
  • Network accessline 509
    separate slash commands. Pass `--no-connect` to register the connector without authenticating, then run `npx @agent-native/core@latest connect https://plan.agent-native.com --client all`
  • Network accessline 510
    authenticating, then run `npx @agent-native/core@latest connect https://plan.agent-native.com --client all` whenever you are ready, or choose a narrower `--client`. Auth and MCP tool
  • Network accessline 510
    authenticating, then run `npx @agent-native/core@latest connect https://plan.agent-native.com --client all` whenever you are ready, or choose a narrower `--client`. Auth and MCP tool
  • Network accessline 511
    `npx @agent-native/core@latest connect https://plan.agent-native.com --client all`
  • Network accessline 511
    `npx @agent-native/core@latest connect https://plan.agent-native.com --client all`
  • Network accessline 535
    Hosted default: connect `https://plan.agent-native.com/mcp`. Do
  • Network accessline 535
    Hosted default: connect `https://plan.agent-native.com/mcp`. Do

skills/visual-plan/references/connection.md

  • Network accessline 34
    - **Codex / Codex Desktop:** run   `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 34
    - **Codex / Codex Desktop:** run   `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 35
    `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 35
    `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`

skills/visual-plan/references/local-files.md

  • Network accessline 76
    is `false`, verify used the offline lint because the app URL was remote or the   local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on
  • Network accessline 76
    is `false`, verify used the offline lint because the app URL was remote or the   local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on
  • Network accessline 77
    local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the
  • Network accessline 77
    local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the
  • Network accessline 78
    `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the   concrete validation error.
  • Network accessline 78
    `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the   concrete validation error.

skills/visual-recap/README.md

  • Network accessline 21
    Visual recaps are MDX, customizable with your own components, and viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan).
  • Network accessline 21
    Visual recaps are MDX, customizable with your own components, and viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan).
  • Network accessline 22
    the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan).
  • Network accessline 22
    the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan).
  • Network accessline 24
    MDX locally, starts a localhost bridge, and opens the hosted Plan UI with no sharing. [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 24
    MDX locally, starts a localhost bridge, and opens the hosted Plan UI with no sharing. [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 26
    [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 26
    [Source here](https://github.com/BuilderIO/agent-native/).
  • Network accessline 76
    can automatically generate a visual recap from every pull request diff. Reviewers open the generated PR recap in the [Plan app review surface](https://www.agent-native.com/docs/template-plan)
  • Network accessline 76
    can automatically generate a visual recap from every pull request diff. Reviewers open the generated PR recap in the [Plan app review surface](https://www.agent-native.com/docs/template-plan)
  • Network accessline 77
    Reviewers open the generated PR recap in the [Plan app review surface](https://www.agent-native.com/docs/template-plan) before moving into raw GitHub diff view.
  • Network accessline 77
    Reviewers open the generated PR recap in the [Plan app review surface](https://www.agent-native.com/docs/template-plan) before moving into raw GitHub diff view.
  • Network accessline 78
    [Plan app review surface](https://www.agent-native.com/docs/template-plan)
  • Network accessline 78
    [Plan app review surface](https://www.agent-native.com/docs/template-plan)
  • Network accessline 87
    ![Example of a visual recap posted to a PR](https://cdn.builder.io/api/v1/image/assets%2FYJIGb4i01jvw0SRdL5Bt%2Fcf9bac396cf24a4ba976fc331af6fc5d)
  • Network accessline 87
    ![Example of a visual recap posted to a PR](https://cdn.builder.io/api/v1/image/assets%2FYJIGb4i01jvw0SRdL5Bt%2Fcf9bac396cf24a4ba976fc331af6fc5d)
  • Network accessline 102
    The skill expects the [Plan MCP connector](https://www.agent-native.com/docs/template-plan)
  • Network accessline 102
    The skill expects the [Plan MCP connector](https://www.agent-native.com/docs/template-plan)

skills/visual-recap/SKILL.md

  • Network accessline 264
    A recap lives only in the database of the MCP that created it. A separately running local dev server (e.g. `http://localhost:8081`) has its OWN database and
  • Network accessline 264
    A recap lives only in the database of the MCP that created it. A separately running local dev server (e.g. `http://localhost:8081`) has its OWN database and
  • Network accessline 265
    running local dev server (e.g. `http://localhost:8081`) has its OWN database and
  • Network accessline 265
    running local dev server (e.g. `http://localhost:8081`) has its OWN database and
  • Network accessline 273
    `visualUrl` in the returned `plan.mdx` frontmatter, else `url`/`path`    resolved against the MCP server's own origin (for the hosted MCP that is    `https://plan.agent-native.com`). This always points at the database that has
  • Network accessline 273
    `visualUrl` in the returned `plan.mdx` frontmatter, else `url`/`path`    resolved against the MCP server's own origin (for the hosted MCP that is    `https://plan.agent-native.com`). This always points at the database that has
  • Network accessline 274
    resolved against the MCP server's own origin (for the hosted MCP that is    `https://plan.agent-native.com`). This always points at the database that has    the plan.
  • Network accessline 274
    resolved against the MCP server's own origin (for the hosted MCP that is    `https://plan.agent-native.com`). This always points at the database that has    the plan.
  • Network accessline 275
    `https://plan.agent-native.com`). This always points at the database that has
  • Network accessline 275
    `https://plan.agent-native.com`). This always points at the database that has
  • Network accessline 277
    2. Use a `localhost`/dev origin ONLY when the recap was created through a Plan    MCP bound to that same origin — i.e. that MCP's url is    `http://localhost:<port>/mcp`. Creating through the hosted MCP
  • Network accessline 277
    2. Use a `localhost`/dev origin ONLY when the recap was created through a Plan    MCP bound to that same origin — i.e. that MCP's url is    `http://localhost:<port>/mcp`. Creating through the hosted MCP
  • Network accessline 278
    MCP bound to that same origin — i.e. that MCP's url is    `http://localhost:<port>/mcp`. Creating through the hosted MCP    and linking to localhost is the exact mismatch that 404s.
  • Network accessline 278
    MCP bound to that same origin — i.e. that MCP's url is    `http://localhost:<port>/mcp`. Creating through the hosted MCP    and linking to localhost is the exact mismatch that 404s.
  • Network accessline 279
    `http://localhost:<port>/mcp`. Creating through the hosted MCP
  • Network accessline 279
    `http://localhost:<port>/mcp`. Creating through the hosted MCP
  • Network accessline 280
    and linking to localhost is the exact mismatch that 404s. 3. If only a plan id is available, build the MCP origin's absolute URL    (hosted: `https://plan.agent-native.com/plans/<id>`) and say it was inferred.
  • Network accessline 280
    and linking to localhost is the exact mismatch that 404s. 3. If only a plan id is available, build the MCP origin's absolute URL    (hosted: `https://plan.agent-native.com/plans/<id>`) and say it was inferred.
  • Network accessline 282
    (hosted: `https://plan.agent-native.com/plans/<id>`) and say it was inferred.
  • Network accessline 282
    (hosted: `https://plan.agent-native.com/plans/<id>`) and say it was inferred.
  • Network accessline 285
    hosted MCP, say so plainly: the local dev server cannot see it. To view a recap on localhost (e.g. to exercise un-deployed local renderer changes), they must connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and
  • Network accessline 285
    hosted MCP, say so plainly: the local dev server cannot see it. To view a recap on localhost (e.g. to exercise un-deployed local renderer changes), they must connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and
  • Network accessline 286
    on localhost (e.g. to exercise un-deployed local renderer changes), they must connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and re-create the recap through it so it lands in the local database; offer to do
  • Network accessline 286
    on localhost (e.g. to exercise un-deployed local renderer changes), they must connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and re-create the recap through it so it lands in the local database; offer to do
  • Network accessline 287
    connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and
  • Network accessline 287
    connect a LOCAL Plan MCP (`http://localhost:<port>/mcp`) and

skills/visual-recap/references/connection.md

  • Network accessline 34
    - **Codex / Codex Desktop:** run   `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 34
    - **Codex / Codex Desktop:** run   `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 35
    `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`
  • Network accessline 35
    `npx -y @agent-native/core@latest reconnect https://plan.agent-native.com --client codex`

skills/visual-recap/references/local-files.md

  • Network accessline 76
    is `false`, verify used the offline lint because the app URL was remote or the   local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on
  • Network accessline 76
    is `false`, verify used the offline lint because the app URL was remote or the   local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on
  • Network accessline 77
    local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the
  • Network accessline 77
    local Plan app was unavailable. Run a local Plan app and pass   `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the
  • Network accessline 78
    `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the   concrete validation error.
  • Network accessline 78
    `--app-url http://localhost:8096` for the authoritative check. If the browser hangs on   "Loading plan", fetch the `bridgeUrl` from the verify/serve JSON to read the   concrete validation error.

skills/webmcp/README.md

  • Network accessline 72
    1. Open `Customize` > `Plugins` > `Add marketplace` in Cowork. 2. Enter `BuilderIO/skills` (or `https://github.com/BuilderIO/skills`).
  • Network accessline 72
    1. Open `Customize` > `Plugins` > `Add marketplace` in Cowork. 2. Enter `BuilderIO/skills` (or `https://github.com/BuilderIO/skills`).
  • Network accessline 73
    2. Enter `BuilderIO/skills` (or `https://github.com/BuilderIO/skills`).
  • Network accessline 73
    2. Enter `BuilderIO/skills` (or `https://github.com/BuilderIO/skills`).

skills/webmcp/SKILL.md

  • Network accessline 48
    - Accept a full URL or hostname; prepend `https://` when the scheme is missing
  • Network accessline 48
    - Accept a full URL or hostname; prepend `https://` when the scheme is missing

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: 306 facts

and 256 more

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexCompatibleStandard SKILL.md skills/agent-watchdog/SKILL.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md skills/agent-watchdog/SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md skills/agent-watchdog/SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md skills/agent-watchdog/SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md skills/agent-watchdog/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.

You might also use

obra/superpowers

An agentic skills framework & software development methodology that works.

313.4kGitHub stars

addyosmani/agent-skills

Production-grade engineering skills for AI coding agents.

113.6kGitHub stars

kepano/obsidian-skills

Agent skills for Obsidian. Teach your agent to use Obsidian CLI and open formats including Markdown, Bases, JSON Canvas.

50.9kGitHub stars

anthropics/claude-plugins-official

Official, Anthropic-managed directory of high quality Claude Code Plugins.

46.3kGitHub stars

mksglu/context-mode

Free, local plugin that keeps tool output out of your coding agent's context window. Sandbox and SQLite FTS5 search, 99% fewer bytes in context. Works with Claude Code, Codex, Pi,…

33.6kGitHub stars

iOfficeAI/OfficeCLI

OfficeCLI is the first and best Office suite purpose-built for AI agents to read, edit, and automate Word, Excel, and PowerPoint files. Free, open-source, single binary, no Offic…

33.1kGitHub stars

Explore alternatives and comparisons

Add this badge to your README

It updates every night and links to this page.

BuilderIO/skills: #292 on hotskills, 4.6k stars

[![BuilderIO/skills on hotskills](https://hotskills.dev/badge/builderio-skills.svg)](https://hotskills.dev/skill/builderio-skills/)