agentleFS
0kim0bos0kim0bos/yam/AGENTS.md

yam

AGENTS.md1 starsChanged 14 days agoMITView on GitHub

Before commit, push, CI approval, or deployment, run the configured yam security check gate for the exact stage and record its Security check block. Use references/security-check.md; preserve every SEC explanation, evidence reference, and unresolved status. Missing required evidence or a note-write failure blocks the stage. Local hooks are bypassable; enforce protected merge/deployment with a trusted CI policy and required checks. Never present a validated supplied report as an independent

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository6 other files it gives its agents.

The file

Show its rules and text

We split this AGENTS.md into the rules it gives an agent

20 rules. About 0 are common, 17 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈17

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start with clear momentum, then deepen deliberately when scope, risk, learning value, or user intent calls for it.

    line 4
  2. Distinctive:

    Treat token economy as part of quality.

    line 5
  3. Distinctive:

    Read only the context needed for the requested change.

    line 7
  4. Distinctive:

    Check local project fit before changing code, even for small work.

    line 8
  5. Distinctive:

    Prefer existing project patterns over new abstractions.

    line 9
  6. Distinctive:

    Verify at the lightest level that honestly supports the claim, then escalate verification with the L0-L5 Verification Ladder when the …

    line 11
  7. Distinctive:

    Do not run broad tests for tiny changes, but do not under-verify risky, security-sensitive, or public work.

    line 13
  8. Distinctive:

    Do not use teams, orchestration, structured proof, or tmux unless risk, user intent, learning value, or an approved route justifies it.

    line 14
  9. Distinctive:

    When code, config, release metadata, documentation, or project artifacts changed, include a Study Note even if no yam skill was explicitly …

    line 18
  10. Distinctive:

    Use external references deliberately, preserve source clarity in research/proof notes, and rework good ideas into yam style.

    line 20
  11. Distinctive:

    Treat security as the first project lens: identify sensitive surfaces early and guide safer defaults before convenience.

    line 21
  12. Distinctive:

    Keep final reports short and concrete, but always include remaining tasks or state that none are meaningful.

    line 22
  13. Very specific:

    When relevant, include architecture hygiene in the Study Note: avoid dumping unrelated logic into page.tsx, one-off component CSS into …

    line 23
  14. Very specific:

    When the user says exactly yam 최신 업데이트 사항 반영해줘 in a chat for this project, treat that phrase as explicit authorization to check and apply …

    line 33
  15. Distinctive:

    Run these commands sequentially from the yam project directory:

    line 35
  16. Distinctive:

    Stop at the first failed command, report the blocker, and do not claim that later steps ran.

    line 45
  17. Distinctive:

    Treat that exact phrase as one-time authorization for this run only.

    line 46
  18. Very specific:

    Never edit .codex/plugins/cache directly.

    line 47
  19. Distinctive:

    After all four commands pass, summarize every component receipt, report the installed yam version and skill status, then tell the user that …

    line 48
  20. Distinctive:

    Do not restart Codex automatically.

    line 49

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

yam Core Rules

  1. Direction before execution.
  2. Start with clear momentum, then deepen deliberately when scope, risk, learning value, or user intent calls for it.
  3. Treat token economy as part of quality.
  4. Reuse project context packs before broad reading.
  5. Read only the context needed for the requested change.
  6. Check local project fit before changing code, even for small work.
  7. Prefer existing project patterns over new abstractions.
  8. Make the smallest useful change.
  9. Verify at the lightest level that honestly supports the claim, then escalate verification with the L0-L5 Verification Ladder when the claim, scope, risk, public release, runtime, UI completion, auth, DB, broad refactor, learning value, or user intent is larger.
  10. Report skipped, partial, blocked, or assumed verification plainly.
  11. Do not run broad tests for tiny changes, but do not under-verify risky, security-sensitive, or public work.
  12. Do not use teams, orchestration, structured proof, or tmux unless risk, user intent, learning value, or an approved route justifies it.
  13. Escalate risky work by asking or recommending, not by silently switching modes.
  14. For UI work, inspect real states and responsive behavior when feasible.
  15. For long-running processes, do not claim cleanup unless process exit is checked.
  16. When code, config, release metadata, documentation, or project artifacts changed, include a Study Note even if no yam skill was explicitly invoked.
  17. Explain useful change insight for non-specialists: what the touched code or artifact does, what role it has, where it runs or is read, what was wrong or missing, what changed before/after, what behavior is expected, one useful syntax/structure insight, what was verified, and what remains uncertain.
  18. Use external references deliberately, preserve source clarity in research/proof notes, and rework good ideas into yam style.
  19. Treat security as the first project lens: identify sensitive surfaces early and guide safer defaults before convenience.
  20. Keep final reports short and concrete, but always include remaining tasks or state that none are meaningful.
  21. When relevant, include architecture hygiene in the Study Note: avoid dumping unrelated logic into page.tsx, one-off component CSS into global.css, or structured product data into broad DB jsonb blobs when existing project boundaries, scoped styles, typed tables, constraints, or indexes fit better.
  22. When an artifact changed, put Next step immediately after Study Note. Ordinary advisory work may use a short concrete action or state that none remains. For risky changes, explicit strict reports and structured receipts, build it from a quick whole-process scan with current situation, forward outlook, a concrete critical opinion, improvement recommendations, and ordered fix-first then planned actions with evidence and ownership. Keep report validation separate from actual security verification.
  23. Work on the current branch by default. Create or switch branches, create worktrees, or create pull requests only when the user explicitly requests that action or an applicable project rule explicitly requires it. A commit/push request, release version bump, route selection, branch-name prefix, or branch protection alone is not authorization. If protection blocks the requested operation, report the blocker and ask for direction; do not silently choose a PR workflow, bypass checks, or change protection settings.

Security check gate

Before commit, push, CI approval, or deployment, run the configured yam security check gate for the exact stage and record its Security check block. Use references/security-check.md; preserve every SEC explanation, evidence reference, and unresolved status. Missing required evidence or a note-write failure blocks the stage. Local hooks are bypassable; enforce protected merge/deployment with a trusted CI policy and required checks. Never present a validated supplied report as an independent vulnerability scan.

Latest Update Trigger

When the user says exactly yam 최신 업데이트 사항 반영해줘 in a chat for this project, treat that phrase as explicit authorization to check and apply reviewed updates for yam, Scrapling, and Insane Search.

Run these commands sequentially from the yam project directory:

yam update check --json
yam update apply --all --json
yam version
yam status
  • Request any environment approval required for network access or writes outside the workspace.
  • Stop at the first failed command, report the blocker, and do not claim that later steps ran.
  • Treat that exact phrase as one-time authorization for this run only; do not create an unattended updater, hook, or automation.
  • Never edit .codex/plugins/cache directly. If Insane Search cannot be updated in place through the official Codex CLI, report manual_plugin_update_required and stop.
  • After all four commands pass, summarize every component receipt, report the installed yam version and skill status, then tell the user that Codex can be restarted to reload updated skills/plugins.
  • Do not restart Codex automatically.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.