agents-md
Audits and edits agent instruction files, verifies repository commands, and connects AGENTS.md and CLAUDE.md without duplicate sources. Use when asked to "improve my AGENTS.md", "write a CLAUDE.md", or make instructions work across agents. For SKILL.md use agent-skills-creator.
mblode · AI review on file
Installation
Read the source and dependencies before running a command. Copying does not install or execute anything.
npx skills add mblode/agent-skills --skill agents-mdAbout this skill
name: agents-md description: Audits and edits agent instruction files, verifies repository commands, and connects AGENTS.md and CLAUDE.md without duplicate sources. Use when asked to "improve my AGENTS.md", "write a CLAUDE.md", or make instructions work across agents. For SKILL.md use agent-skills-creator.
AGENTS.md Setup and Audit
- IS: wiring a repo so every agent tool in use reads the same rules, then auditing, scoring, refactoring, and writing the AGENTS.md / CLAUDE.md / CLAUDE.local.md files agents load at session start.
- IS NOT: authoring SKILL.md files (use
agent-skills-creator), project docs or READMEs (usedocs-writingorreadme-creator), or mining session history (use the externalcadence-adviseskill where installed; this skill audits the file as-is).
AGENTS.md files are execution contracts, not knowledge bases. Two tests catch the two ways a line fails.
- Dead weight: "Would removing this cause the agent to make a mistake?" If no, cut it; bloat makes agents ignore the rules that matter.
- Harmful precision: "Is this wrong on any plausible task in this repo?" A prohibition that is wrong one task in ten is still obeyed on that task, and the agent cannot tell that this is the exception. State the outcome you want and let the surrounding code pick the path.
NEVER write commentsbecomesmatch the comment density of the file you are editing: shorter, no exception list to maintain, and correct in a densely commented file without being told.
Absolutes still earn their place for safety, data loss, format contracts, and rules this repo's agents have actually been observed to break.
AGENTS.md is the tool-agnostic source of truth: Codex, Cursor, Copilot, and the rest of the agents.md list read it natively. Claude Code is the exception. It reads CLAUDE.md, not AGENTS.md, so a repo that uses Claude Code needs a CLAUDE.md whose first line is @AGENTS.md (or a CLAUDE.md -> AGENTS.md symlink when there are no Claude-only additions). If only a CLAUDE.md exists and another tool is in use, git mv CLAUDE.md AGENTS.md and add the pointer file; a copy of either drifts.
Choose a Mode
- Repo has no agent instructions, or targets a tool it is not wired for -> Setup:
references/project-setup.mddecides which files exist and which tool reads each one, then Writing From Scratch below fills the root file, then Audit scores it. - A file exists and the question is quality -> Audit, below.
- A file exists and is bloated, stale, or scored badly -> Refactor,
references/refactor-workflow.md.
Reference Files
| File | Read when |
|---|---|
references/project-setup.md | Setting a repo up for Claude Code, Codex, and Cursor; deciding which files exist and what each tool actually loads |
references/quick-checklist.md | Every audit; default 12-check triage |
references/quality-criteria.md | Quick audit fails, file is high-risk, or full scoring requested |
references/refactor-workflow.md | File is bloated (root over ~150 lines), stale, or below target |
references/root-content-guidance.md | Deciding what stays in root and where moved content goes so it still loads |
references/templates.md | Drafting or rebuilding a file from scratch |
Writing From Scratch
The content step of Setup, and the whole job when the repo is already wired and only the file is missing. Skip the audit. Gather real commands from the manifest (package.json, Makefile, CI config), pick a skeleton from references/templates.md, fill it with verified commands and known gotchas, then validate against references/quick-checklist.md before delivering.
Audit Workflow
Copy this checklist to track progress:
Audit Progress:
- [ ] Step 1: Discover files
- [ ] Step 2: Select audit mode (quick or full)
- [ ] Step 3: Run audit and score
- [ ] Step 4: Report findings with score table
- [ ] Step 5: Propose minimal diffs
- [ ] Step 6: Validate changes
- [ ] Step 7: Apply and report before/after scores
Step 1: Discover files
find . \( -name "AGENTS.md" -o -name "AGENTS.override.md" -o -name "CLAUDE.md" -o -name "CLAUDE.local.md" \) -not -path "*/node_modules/*" 2>/dev/null | sort
ls -la CLAUDE.md .claude/rules .cursor/rules 2>/dev/null
Also check ~/.claude/CLAUDE.md and ~/.codex/AGENTS.md; both load in every repo. ls -la CLAUDE.md tells you whether it is a symlink, an @AGENTS.md pointer, or a second copy, and a copy is a finding on its own. For monorepos, include workspace-level files. Audit each level independently: root holds universal rules, child files hold directory-specific rules (see the placement hierarchy in references/root-content-guidance.md).
Step 2: Select audit mode
- Quick (default): 12 checks from
references/quick-checklist.md, target >= 10/12. - Full: 49 checks from
references/quality-criteria.md, target >= 91% of applicable points (grade A). Use when the quick audit fails, the file gates a high-risk repo, or full scoring is requested.
Step 3: Run audit and score
Score each root file independently; exclude N/A checks from the denominator.
Step 4: Report findings
Output a concise report before any edits:
## AGENTS.md Audit Report
| File | Mode | Score | Grade | Key Issues |
|------|------|-------|-------|------------|
| ./AGENTS.md | Quick | 6/10 | Fail | Missing test command, stale path, doc-heavy section |
Every issue in the table must map to a Step 5 diff; no vague findings.
Step 5: Propose minimal diffs
In priority order:
- Fix broken or stale commands; bugs, not style.
- Remove generic, duplicate, or obsolete guidance, restatements of what the harness already does, and facts auto-memory owns.
- Rewrite blanket prohibitions as the outcome they were protecting; keep the absolute only where the harmful-precision test clears it.
- Move detail needed in fewer than ~30% of tasks to a location that loads on demand: a nest