Not in the top 1,000 by GitHub stars
agfnow/agentflow-v8
File-logged conversation protocol and on-demand development pipeline
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/agentflow/SKILL.md
Week of Nov 2, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 0bda4312bf3e
No earlier version was observed for comparison.
Verified commit 0bda4312bf3e781d22468923ce35e9d119378b6f
git clone --filter=blob:none --no-checkout https://github.com/agfnow/agentflow-v8.git .hotskills/agfnow-agentflow-v8 && git -C .hotskills/agfnow-agentflow-v8 checkout --detach 0bda4312bf3e781d22468923ce35e9d119378b6f
# Copy each selected skill directory from .hotskills/agfnow-agentflow-v8 into .claude/skills/<skill-name>/Heat
1.1- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
- agentflow
skills/agentflow
How it works
agentflow
When your agent uses it
File logs, Git evidence and optional development. Triggered by godev/devlog/ag/agentflow/fast-lane/skip-ag.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
7,256 tokensWhen the skill is activated
Reference documents
61,221 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Start here
- Every message after startup
- Load rules only when triggered
- Task artifact locations
- Scope and evidence
- Writing styles protocol
- Progress records
- Completing a round
- Settings
- Final safety
Scripts executed separately
scripts/ag-settings.jsscripts/ag-settings.test.jsscripts/agf.jsscripts/agf.test.jsscripts/alignment.test.jsscripts/ask-names.test.jsscripts/branch-safety-terminal.test.jsscripts/claude-live-journey.jsscripts/close-language-journey.jsscripts/closed-round.jsscripts/closed-round.test.jsscripts/codex-live-journey.jsscripts/completion-cleanup-integration.test.jsscripts/completion-cleanup.jsscripts/completion-cleanup.test.jsscripts/completion-context.jsscripts/completion-context.test.jsscripts/completion-record.jsscripts/completion-record.test.jsscripts/cross-check-plan.jsscripts/cross-check-plan.test.jsscripts/default-branch.jsscripts/default-branch.test.jsscripts/delegation-route.jsscripts/delegation-route.test.jsscripts/devlog-guard.jsscripts/devlog-guard.test.jsscripts/external-runner.jsscripts/external-runner.test.jsscripts/fast-lane.jsscripts/fast-lane.test.jsscripts/fixtures/external-worker.jsscripts/fixtures/notebook-owner.jsscripts/fixtures/temp-directory.jsscripts/fixtures/temp-directory.test.jsscripts/git-optional.test.jsscripts/install-hook.jsscripts/install-hook.test.jsscripts/language-contract.test.jsscripts/local-time.jsscripts/local-time.test.jsscripts/log-controls-journey.jsscripts/log-controls.test.jsscripts/long-round-hook-journey.jsscripts/looper-live-gate.jsscripts/looper-live-gate.test.jsscripts/looper.jsscripts/looper.test.jsscripts/no-ag-closeout.test.jsscripts/no-ag-hook-journey.jsscripts/no-ag-hook.test.jsscripts/notebook-boundary-journey.test.jsscripts/notebook-compact.jsscripts/notebook-compact.test.jsscripts/notebook-looper-owner.test.jsscripts/notebook-owner-first-stream.test.jsscripts/notebook-owner-recovery.test.jsscripts/notebook-owner.jsscripts/notebook-owner.test.jsscripts/notebook-write.jsscripts/notebook-write.test.jsscripts/owner-control-text.jsscripts/ownership-controls-journey.jsscripts/ownership-controls.test.jsscripts/portable-host.test.jsscripts/process-tree.jsscripts/process-tree.test.jsscripts/prompt-compression.test.jsscripts/queue-contract.jsscripts/queue-contract.test.jsscripts/queue-routing.test.jsscripts/queue-transports.test.jsscripts/release.test.jsscripts/reply-identity-journey.pyscripts/reply-identity.jsscripts/reply-identity.test.jsscripts/repository-state.jsscripts/resume-intake.jsscripts/resume-intake.test.jsscripts/review-only-journey.jsscripts/review-only.test.jsscripts/review-policy.test.jsscripts/round-linter.jsscripts/round-linter.test.jsscripts/setup.jsscripts/setup.test.jsscripts/skills-audit-journey.pyscripts/skills-audit.jsscripts/skills-audit.test.jsscripts/start-journey.test.jsscripts/stop-hook-recovery.test.jsscripts/stop-hook.jsscripts/stop-hook.test.jsscripts/stream-cleanup.jsscripts/streams-off.test.jsscripts/suite-evidence.jsscripts/suite-evidence.test.jsscripts/terminal-preflight.jsscripts/terminal-preflight.test.jsscripts/terminal.test.jsscripts/threeways-tier-journey.jsscripts/tracker-contract.jsscripts/transport-integration.test.jsscripts/transport-journey.test.js
Files
../../.claude-plugin/marketplace.jsonasset279 bytes../../.claude-plugin/plugin.jsonasset219 bytesSKILL.mdcore36,895 bytesSKILL.mdcore36,895 bytesSKILL.mdcore36,895 bytesdocs/AG_GUIDE.mddoc34,780 bytesdocs/AG_GUIDE.zh-tw.mddoc32,443 bytesdocs/agent-brief.mddoc15,098 bytesdocs/incidents-log.mddoc52,658 bytesdocs/skill-editing-and-compression.mddoc2,278 bytesreferences/advisors/acceptance.mddoc4,926 bytesreferences/advisors/codewalk.mddoc2,847 bytesreferences/advisors/explore.mddoc2,425 bytesreferences/advisors/learn.mddoc2,222 bytesreferences/advisors/requirements.mddoc4,253 bytesreferences/advisors/security-scan.mddoc2,597 bytesreferences/advisors/spec.mddoc3,754 bytesreferences/advisors/spike.mddoc2,163 bytesreferences/ag.mddoc19,119 bytesreferences/closeout.mddoc22,642 bytesreferences/delegation.mddoc15,594 bytesreferences/fast-lane.mddoc2,971 bytesreferences/looper.mddoc8,563 bytesreferences/mixed-requests.mddoc3,616 bytesreferences/progress.mddoc3,230 bytesreferences/skill-conflicts.mddoc3,157 bytesreferences/skip-ag.mddoc1,532 bytesreferences/streams.mddoc11,445 bytesreferences/writing.mddoc5,429 bytesscripts/README.mddoc21,836 bytesscripts/ag-settings.jsruns116,011 bytesscripts/ag-settings.test.jsruns111,133 bytesscripts/agf.jsruns132,973 bytesscripts/agf.test.jsruns175,922 bytesscripts/alignment.test.jsruns16,682 bytesscripts/ask-names.test.jsruns6,611 bytesscripts/branch-safety-terminal.test.jsruns3,520 bytesscripts/claude-live-journey.jsruns9,362 bytesscripts/close-language-journey.jsruns3,760 bytesscripts/closed-round.jsruns1,695 bytesscripts/closed-round.test.jsruns2,802 bytesscripts/codex-live-journey.jsruns8,808 bytesscripts/completion-cleanup-integration.test.jsruns6,741 bytesscripts/completion-cleanup.jsruns24,550 bytesscripts/completion-cleanup.test.jsruns8,979 bytesscripts/completion-context.jsruns29,335 bytesscripts/completion-context.test.jsruns42,788 bytesscripts/completion-record.jsruns25,116 bytesscripts/completion-record.test.jsruns22,706 bytesscripts/cross-check-plan.jsruns7,934 bytesscripts/cross-check-plan.test.jsruns8,033 bytesscripts/default-branch.jsruns1,433 bytesscripts/default-branch.test.jsruns2,646 bytesscripts/delegation-route.jsruns47,673 bytesscripts/delegation-route.test.jsruns32,339 bytesscripts/devlog-guard.jsruns2,960 bytesscripts/devlog-guard.test.jsruns5,477 bytesscripts/external-runner.jsruns27,139 bytesscripts/external-runner.test.jsruns21,015 bytesscripts/fast-lane.jsruns1,420 bytesscripts/fast-lane.test.jsruns15,507 bytesscripts/fixtures/external-worker.jsruns2,453 bytesscripts/fixtures/notebook-owner.jsruns1,861 bytesscripts/fixtures/temp-directory.jsruns544 bytesscripts/fixtures/temp-directory.test.jsruns1,225 bytesscripts/git-optional.test.jsruns17,764 bytesscripts/install-hook.jsruns15,468 bytesscripts/install-hook.test.jsruns21,594 bytesscripts/language-contract.test.jsruns1,503 bytesscripts/local-time.jsruns2,138 bytesscripts/local-time.test.jsruns1,326 bytesscripts/log-controls-journey.jsruns6,541 bytesscripts/log-controls.test.jsruns14,382 bytesscripts/long-round-hook-journey.jsruns3,668 bytesscripts/looper-live-gate.jsruns8,826 bytesscripts/looper-live-gate.test.jsruns3,451 bytesscripts/looper.jsruns96,983 bytesscripts/looper.test.jsruns140,642 bytesscripts/no-ag-closeout.test.jsruns7,400 bytesscripts/no-ag-hook-journey.jsruns3,006 bytesscripts/no-ag-hook.test.jsruns6,806 bytesscripts/notebook-boundary-journey.test.jsruns6,225 bytesscripts/notebook-compact.jsruns9,907 bytesscripts/notebook-compact.test.jsruns17,011 bytesscripts/notebook-looper-owner.test.jsruns15,977 bytesscripts/notebook-owner-first-stream.test.jsruns5,435 bytesscripts/notebook-owner-recovery.test.jsruns5,691 bytesscripts/notebook-owner.jsruns23,206 bytesscripts/notebook-owner.test.jsruns29,880 bytesscripts/notebook-write.jsruns62,326 bytesscripts/notebook-write.test.jsruns63,424 bytesscripts/owner-control-text.jsruns340 bytesscripts/ownership-controls-journey.jsruns4,089 bytesscripts/ownership-controls.test.jsruns9,836 bytesscripts/portable-host.test.jsruns7,502 bytesscripts/process-tree.jsruns3,529 bytesscripts/process-tree.test.jsruns2,087 bytesscripts/prompt-compression.test.jsruns8,149 bytesscripts/queue-contract.jsruns62,734 bytesscripts/queue-contract.test.jsruns16,153 bytesscripts/queue-routing.test.jsruns10,762 bytesscripts/queue-transports.test.jsruns11,212 bytesscripts/release.test.jsruns984 bytesscripts/reply-identity-journey.pyruns3,936 bytesscripts/reply-identity.jsruns4,910 bytesscripts/reply-identity.test.jsruns6,102 bytesscripts/repository-state.jsruns2,056 bytesscripts/resume-intake.jsruns16,189 bytesscripts/resume-intake.test.jsruns12,970 bytesscripts/review-only-journey.jsruns6,203 bytesscripts/review-only.test.jsruns13,704 bytesscripts/review-policy.test.jsruns12,145 bytesscripts/round-linter.jsruns212,892 bytesscripts/round-linter.test.jsruns199,178 bytesscripts/setup.jsruns18,977 bytesscripts/setup.test.jsruns25,246 bytesscripts/skills-audit-journey.pyruns2,484 bytesscripts/skills-audit.jsruns4,740 bytesscripts/skills-audit.test.jsruns2,490 bytesscripts/start-journey.test.jsruns23,096 bytesscripts/stop-hook-recovery.test.jsruns9,420 bytesscripts/stop-hook.jsruns11,509 bytesscripts/stop-hook.test.jsruns36,540 bytesscripts/stream-cleanup.jsruns5,618 bytesscripts/streams-off.test.jsruns1,444 bytesscripts/suite-evidence.jsruns10,065 bytesscripts/suite-evidence.test.jsruns6,367 bytesscripts/terminal-preflight.jsruns2,301 bytesscripts/terminal-preflight.test.jsruns2,637 bytesscripts/terminal.test.jsruns50,709 bytesscripts/threeways-tier-journey.jsruns1,997 bytesscripts/tracker-contract.jsruns5,448 bytesscripts/transport-integration.test.jsruns8,618 bytesscripts/transport-journey.test.jsruns8,003 bytes
Read SKILL.md
# Agentflow v8.4.15
Agentflow keeps owner conversation and live recovery in a configured notebook; advanced rules load only on demand.
`stream(branch/worktree)` is an owner/session feature workspace, usually a branch/worktree with adjacent notebook and configuration. Non-code notebook-only requests may use a stream file and root pointer without branching, even with Git. Ordinary notebooks support no-Git folders; `new-feature:` requires Git. External workers use disposable no-remote clones; internal/host work uses actual tools and recorded ownership.
For releases, update version and heading together: major/minor/patch for breaking changes/features/fixes. This file supplies the release version. Follow repository release instructions; task commits are not releases. For a correction to an already-published release, first fetch and inspect the public branch and treat its current version and files as authoritative; never infer them from a private development checkout. A same-version public correction must preserve newer public files and verify the exact remote commit after pushing.
Incident citations explain failures. Approved redesigns may replace obsolete remedies; preserve hazard checks and history.
## Start here
Already-launched looper workers follow their supplied plan directly, not this host startup/closeout protocol.
View the complete source on GitHubSkills included
- agentflow
Repository facts
- License
- Apache-2.0
- Created
- Sep 4, 2026
- Last pushed
- Oct 23, 2026
- Forks
- 44
- Open issues
- 4
- Project activity
- Active
Before you install
Local scriptsAutomated signals, not a guarantee
Checked Nov 6, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 2 notes
skills/agentflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/agentflow/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 22 network referencesin 4 files: agentflow/scripts/agf.test.js:2502 agentflow/docs/AG_GUIDE.md:5 agentflow/docs/AG_GUIDE.zh-tw.md:5 and 1 more
- No hidden or obfuscated text
- 5 instruction override signalsin 1 file: agentflow/scripts/claude-live-journey.js:79
- 104 executable scriptsin 104 files: agentflow/scripts/ag-settings.js:1 agentflow/scripts/ag-settings.test.js:1 agentflow/scripts/agf.js:1 and 101 more
- 8 references to secretsin 7 files: agentflow/scripts/agf.js:1084 agentflow/scripts/ag-settings.js:963 agentflow/scripts/agf.test.js:1572 and 4 more
- No requests to vote in Hot
Show all 139 signals by file
skills/agentflow/docs/AG_GUIDE.md
- Network accessline 5
[Watch the Agentflow introduction on YouTube](https://youtu.be/0dp_HnqX0ms). - Network accessline 29
Choose the assistant you use and whether to install for this project or all your projects. Then start a new assistant session so it can load the skill. Claude Code also has a plugin installation route; see the [public README](https://github - Network accessline 31
For Codex, the Agentflow maintainer recommends **`gpt-6.1-sol/medium` as the most stable coordinator choice in their use**. This means model `gpt-6.1-sol` with reasoning effort `medium`, running the main conversation and coordinating work. - Network accessline 43
- Install the applications with `brew install pi-coding-agent can1357/tap/omp`, then run `pi` or `omp` from your project. These are the [Pi Homebrew formula](https://formulae.brew.sh/formula/pi-coding-agent) and [OMP's documented tap](https - Network accessline 66
The current CLI updates by **installed skill name**. Use `agentflow` here; `agfnow/agentflow` is the source used by `skills add`. The source-style command `npx skills update agfnow/agentflow` is not the current name-based syntax. Add `-g` f
skills/agentflow/docs/AG_GUIDE.zh-tw.md
- Network accessline 5
[在 YouTube 觀看 Agentflow 介紹影片](https://youtu.be/0dp_HnqX0ms)。 - Network accessline 29
依畫面選擇你使用的 AI 助理,以及只安裝在這個專案,還是供所有專案使用。安裝後開啟新的助理對話,讓它載入 skill。Claude Code 也能透過外掛安裝,方式請看[公開版 README](https://github.com/agfnow/agentflow#install)。 - Network accessline 31
如果你使用 Codex,Agentflow 維護者依自己的使用經驗,建議選擇 **`gpt-6.1-sol/medium` 作為表現最穩定的控場模型**。也就是模型選 `gpt-6.1-sol`,推理程度選 `medium`,由它負責和你對話、安排工作。這是本專案的使用建議,仍可依任務與帳號可用的模型調整。[OpenAI 官方模型文件](https://developers.openai.com/api/docs/models/gpt-6.1-sol)也確認這個模型支援 `m - Network accessline 49
目前的 CLI 是依**已安裝的 skill 名稱**更新,所以這裡填 `agentflow`;`agfnow/agentflow` 是 `skills add` 使用的安裝來源。`npx skills update agfnow/agentflow` 這種填入來源的寫法,不是目前依名稱更新的語法。全域安裝可加上 `-g`;專案內安裝則在該專案資料夾執行,並加上 `-p`。詳見 [Skills 更新說明](https://github.com/vercel-labs/skil
skills/agentflow/scripts/ag-settings.js
skills/agentflow/scripts/ag-settings.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/agf.js
skills/agentflow/scripts/agf.test.js
- Executable scriptline 1
Executable file - Access to secretsline 1572
[Credential-related source omitted] - Network accessline 2502
const esc = String.fromCharCode(27) const bell = String.fromCharCode(7) const raw = esc + '[31mremote ' + esc + ']0;title' + bell + 'https://user:secret@example.test/repo' + esc + '[0m' - Network accessline 2503
const bell = String.fromCharCode(7) const raw = esc + '[31mremote ' + esc + ']0;title' + bell + 'https://user:secret@example.test/repo' + esc + '[0m' assert.equal(agf.sanitize_diagnostic(raw), 'remote https://[redacted]@example.test/repo' - Network accessline 2504
const raw = esc + '[31mremote ' + esc + ']0;title' + bell + 'https://user:secret@example.test/repo' + esc + '[0m' - Network accessline 2505
assert.equal(agf.sanitize_diagnostic(raw), 'remote https://[redacted]@example.test/repo') - Network accessline 2603
commit_stream_file(run, wt, 'feature.txt', 'feature\n') const hook = path.join(bare, 'hooks', 'pre-receive') fs.writeFileSync(hook, '#!/bin/sh\nprintf \"\\\\033[31mhttps://user:secret@example.test/repo\\\\033]0;title\\\\007\\\\n\" >&2\nex - Network accessline 2604
const hook = path.join(bare, 'hooks', 'pre-receive') fs.writeFileSync(hook, '#!/bin/sh\nprintf \"\\\\033[31mhttps://user:secret@example.test/repo\\\\033]0;title\\\\007\\\\n\" >&2\nexit 1\n', { mode: 0o755 }) const logs = [] - Network accessline 2605
fs.writeFileSync(hook, '#!/bin/sh\nprintf \"\\\\033[31mhttps://user:secret@example.test/repo\\\\033]0;title\\\\007\\\\n\" >&2\nexit 1\n', { mode: 0o755 }) - Network accessline 2886
commit_stream_file(rejected.run, rejected_wt, 'app.txt', 'feature\n') const hook = path.join(rejected.bare, 'hooks', 'pre-receive') fs.writeFileSync(hook, '#!/bin/sh\nwhile read old new ref\ndo\n case "$ref" in\n refs/heads/main) prin - Network accessline 2887
const hook = path.join(rejected.bare, 'hooks', 'pre-receive') fs.writeFileSync(hook, '#!/bin/sh\nwhile read old new ref\ndo\n case "$ref" in\n refs/heads/main) printf "reject https://user:secret@example.test\\\\033[31m\\\\n" >&2; exit - Network accessline 2888
fs.writeFileSync(hook, '#!/bin/sh\nwhile read old new ref\ndo\n case "$ref" in\n refs/heads/main) printf "reject https://user:secret@example.test\\\\033[31m\\\\n" >&2; exit 1 ;;\n esac\ndone\nexit 0\n', { mode: 0o755 })
skills/agentflow/scripts/alignment.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/ask-names.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/branch-safety-terminal.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/claude-live-journey.js
- Executable scriptline 1
#!/usr/bin/env node - Instruction override signalline 79
global spawn_id env if {$resume} { spawn claude --dangerously-skip-permissions --strict-mcp-config --resume $env(JOURNEY_SESSION) - Instruction override signalline 80
if {$resume} { spawn claude --dangerously-skip-permissions --strict-mcp-config --resume $env(JOURNEY_SESSION) } else { - Instruction override signalline 81
spawn claude --dangerously-skip-permissions --strict-mcp-config --resume $env(JOURNEY_SESSION) - Instruction override signalline 82
} else { spawn claude --dangerously-skip-permissions --strict-mcp-config --session-id $env(JOURNEY_SESSION) } - Instruction override signalline 83
spawn claude --dangerously-skip-permissions --strict-mcp-config --session-id $env(JOURNEY_SESSION)
skills/agentflow/scripts/close-language-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/closed-round.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/closed-round.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/codex-live-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/completion-cleanup-integration.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-cleanup.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-cleanup.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-context.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-context.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-record.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/completion-record.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/cross-check-plan.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/cross-check-plan.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/default-branch.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/default-branch.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/delegation-route.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/delegation-route.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/devlog-guard.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/devlog-guard.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/external-runner.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/external-runner.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fast-lane.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fast-lane.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fixtures/external-worker.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fixtures/notebook-owner.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fixtures/temp-directory.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/fixtures/temp-directory.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/git-optional.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/install-hook.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/install-hook.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/language-contract.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/local-time.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/local-time.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/log-controls-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/log-controls.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/long-round-hook-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/looper-live-gate.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/looper-live-gate.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/looper.js
skills/agentflow/scripts/looper.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/no-ag-closeout.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/no-ag-hook-journey.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/no-ag-hook.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-boundary-journey.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-compact.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-compact.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-looper-owner.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-owner-first-stream.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-owner-recovery.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-owner.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-owner.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/notebook-write.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/notebook-write.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/owner-control-text.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/ownership-controls-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/ownership-controls.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/portable-host.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/process-tree.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/process-tree.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/prompt-compression.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/queue-contract.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/queue-contract.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/queue-routing.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/queue-transports.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/release.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/reply-identity-journey.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 34
[Credential-related source omitted] - Network accessline 36
node = shlex.quote(shutil.which('node')) command = f'{node} {shlex.quote(str(writer))} append-reply --notebook devlog.md --ask ' script = f'test -t 0 && test -t 1 && echo PTY_OK\n{command}A-001 --input-stdin < reply.md\n{node} next.js\n{com - Network accessline 37
command = f'{node} {shlex.quote(str(writer))} append-reply --notebook devlog.md --ask ' script = f'test -t 0 && test -t 1 && echo PTY_OK\n{command}A-001 --input-stdin < reply.md\n{node} next.js\n{command}A-002 --input-stdin < reply.md\nresu - Network accessline 38
script = f'test -t 0 && test -t 1 && echo PTY_OK\n{command}A-001 --input-stdin < reply.md\n{node} next.js\n{command}A-002 --input-stdin < reply.md\nresult=$?\ncat devlog.md\necho JOURNEY_EXIT:$result\nexit "$result"\n'
skills/agentflow/scripts/reply-identity.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/reply-identity.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/repository-state.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/resume-intake.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/resume-intake.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/review-only-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/review-only.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/review-policy.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/round-linter.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/round-linter.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/setup.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/setup.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/skills-audit-journey.py
skills/agentflow/scripts/skills-audit.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/skills-audit.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/start-journey.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/stop-hook-recovery.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/stop-hook.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/stop-hook.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/stream-cleanup.js
skills/agentflow/scripts/streams-off.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/suite-evidence.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/suite-evidence.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/terminal-preflight.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/terminal-preflight.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/terminal.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/threeways-tier-journey.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/tracker-contract.js
- Executable scriptline 1
#!/usr/bin/env node
skills/agentflow/scripts/transport-integration.test.js
- Executable scriptline 1
Executable file
skills/agentflow/scripts/transport-journey.test.js
- Executable scriptline 1
Executable file
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.
Signal confidence: 139 facts
- Network access in prose of documentationagentflow/docs/AG_GUIDE.md:5Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.md:29Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.md:31Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.md:43Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.md:66Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.zh-tw.md:5Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.zh-tw.md:29Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.zh-tw.md:31Low confidence
- Network access in prose of documentationagentflow/docs/AG_GUIDE.zh-tw.md:49Low confidence
- Executable script in code of a file that runsagentflow/scripts/ag-settings.js:1High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueagentflow/scripts/ag-settings.js:963Low confidence
- Executable script in code of a testagentflow/scripts/ag-settings.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/agf.js:1High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueagentflow/scripts/agf.js:1084Low confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueagentflow/scripts/agf.js:1225Low confidence
- Executable script in code of a testagentflow/scripts/agf.test.js:1Low confidence
- Access to secrets in code of a test: a variable, placeholder or call, not a stored valueagentflow/scripts/agf.test.js:1572Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2502Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2503Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2504Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2505Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2603Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2604Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2605Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2886Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2887Low confidence
- Network access in code of a testagentflow/scripts/agf.test.js:2888Low confidence
- Executable script in code of a testagentflow/scripts/alignment.test.js:1Low confidence
- Executable script in code of a testagentflow/scripts/ask-names.test.js:1Low confidence
- Executable script in code of a testagentflow/scripts/branch-safety-terminal.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/claude-live-journey.js:1High confidence
- Instruction override signal in code of a file that runsagentflow/scripts/claude-live-journey.js:79Low confidence
- Instruction override signal in code of a file that runsagentflow/scripts/claude-live-journey.js:80Low confidence
- Instruction override signal in code of a file that runsagentflow/scripts/claude-live-journey.js:81Low confidence
- Instruction override signal in code of a file that runsagentflow/scripts/claude-live-journey.js:82Low confidence
- Instruction override signal in code of a file that runsagentflow/scripts/claude-live-journey.js:83Low confidence
- Executable script in code of a file that runsagentflow/scripts/close-language-journey.js:1High confidence
- Executable script in code of a file that runsagentflow/scripts/closed-round.js:1High confidence
- Executable script in code of a testagentflow/scripts/closed-round.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/codex-live-journey.js:1High confidence
- Executable script in code of a testagentflow/scripts/completion-cleanup-integration.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/completion-cleanup.js:1High confidence
- Executable script in code of a testagentflow/scripts/completion-cleanup.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/completion-context.js:1High confidence
- Executable script in code of a testagentflow/scripts/completion-context.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/completion-record.js:1High confidence
- Executable script in code of a testagentflow/scripts/completion-record.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/cross-check-plan.js:1High confidence
- Executable script in code of a testagentflow/scripts/cross-check-plan.test.js:1Low confidence
- Executable script in code of a file that runsagentflow/scripts/default-branch.js:1High confidence
and 89 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md skills/agentflow/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md skills/agentflow/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md skills/agentflow/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md skills/agentflow/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md skills/agentflow/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.