agentleFS
veep2012veep2012/doc_mcp/.agents/skills/monthly-doc-actualizer/SKILL.md

monthly-doc-actualizer

Skill0 starsChanged 3 months agoMITView on GitHub

Run daily documentation freshness checks and trigger full code-first documentation actualization when due. Use when repository docs must be revalidated against the current implementation in `src/`, `config/`, `scripts/`, `tests/`, and related source files.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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:

    Use this skill to enforce periodic documentation refresh with a lightweight daily check and a full monthly actualization when due.

    line 10
  2. Distinctive:

    If code and docs disagree, update docs to match the implemented behavior unless the user explicitly asks for a code change instead.

    line 15
  3. Distinctive:

    Stop after the check.

    line 35
  4. Very specific:

    Run full documentation actualization for documentation/*.md as a code-first review:

    line 39
  5. Very specific:

    read documentation/_documentation-index.md to identify canonical docs by domain

    line 40
  6. Distinctive:

    update the canonical docs to match current code behavior

    line 53
  7. Distinctive:

    remove obsolete claims instead of leaving “future”, “planned”, or stale historical wording in current-state docs

    line 54
  8. Very specific:

    keep documentation/_documentation-index.md aligned for any added or renamed docs

    line 56
  9. Distinctive:

    Do not assume unchanged docs are accurate because they were recently touched.

    line 66
  10. Distinctive:

    If a document is still accurate after review, leave it unchanged and mention that it was checked.

    line 67
  11. Distinctive:

    After successful actualization and validation, mark completion:

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

Monthly Doc Actualizer

Overview

Use this skill to enforce periodic documentation refresh with a lightweight daily check and a full monthly actualization when due.

Core principle:

  • Actualization is not a wording pass.
  • Actualization means reading the current implementation first, identifying canonical docs from documentation/_documentation-index.md, and updating those docs so they describe the code that exists now.
  • If code and docs disagree, update docs to match the implemented behavior unless the user explicitly asks for a code change instead.

State is tracked in:

  • documentation/_documentation_actualization_state.md

Helper script:

  • .agents/skills/monthly-doc-actualizer/scripts/check_due.py

Workflow (Required)

Step 1: Run daily due check

  • Run:
    • python3 .agents/skills/monthly-doc-actualizer/scripts/check_due.py --state-file documentation/_documentation_actualization_state.md --mark-check
  • Read output fields:
    • DUE=true|false
    • LAST_FULL_ACTUALIZATION
    • NEXT_DUE_DATE
    • DAYS_SINCE_FULL

Step 2: If not due

  • Stop after the check.
  • Report that monthly actualization is not required yet.

Step 3: If due

  • Run full documentation actualization for documentation/*.md as a code-first review:
    • read documentation/_documentation-index.md to identify canonical docs by domain
    • inspect current implementation before editing docs:
      • src/
      • config/
      • scripts/
      • tests/
      • any feature-specific source files touched by recent work
    • compare implementation to documentation and classify each mismatch:
      • stale doc text
      • stale examples
      • stale schema, table, or field descriptions
      • stale endpoint, request, response, or validation contract
      • stale test-scenario mapping
    • update the canonical docs to match current code behavior
    • remove obsolete claims instead of leaving “future”, “planned”, or stale historical wording in current-state docs
    • normalize Document Control metadata (Last Updated, Version) in modified docs
    • keep documentation/_documentation-index.md aligned for any added or renamed docs
    • preserve repository doc standards, including one Change Log entry per calendar date

Step 4: Required coverage during a full actualization

  • Review at least these categories when they exist in the repo:
    • API contract docs
    • workflow and DB rule docs
    • ER and data-model docs
    • auth and permission docs
    • test scenario docs
  • Do not assume unchanged docs are accurate because they were recently touched; confirm against current code.
  • If a document is still accurate after review, leave it unchanged and mention that it was checked.

Step 5: Validate

  • Invariant: documentation markdown filenames must be lowercase (no uppercase letters).
  • Run:
    • find documentation -type f -name '*[A-Z]*.md'
  • Expected: no output.
  • Confirm no document has more than one Change Log entry for the same calendar date.
  • Confirm modified docs still satisfy documentation/_documentation_standards.md.

Step 6: Mark completion

  • After successful actualization and validation, mark completion:
    • python3 .agents/skills/monthly-doc-actualizer/scripts/check_due.py --state-file documentation/_documentation_actualization_state.md --mark-full

Output Requirements

Always report:

  • Whether monthly run was due
  • Which implementation areas were reviewed
  • Which documentation files were checked
  • Which files were updated
  • New next due date
  • Any unresolved ambiguity or manual follow-up

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.