agentleFS
fortunto2fortunto2/solo-factory/skills/audit/SKILL.md

solo-audit

Skill18 starsChanged 4 months agoMITView on GitHub

Use when "audit KB", "check frontmatter", "find broken links", "tag cleanup", "knowledge base quality", or docs need health check. Do NOT use for SEO audits (/seo-audit) or code reviews (/review).

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Find all markdown files: Use Glob to find all .md files, excluding common non-content directories: .venv/, node_modules/, .git/, archive/, …

    line 22
  2. Distinctive:

    Link check: Look for broken internal links:

    line 31
  3. Very specific:

    If a link-checking script exists in the project (e.g., scripts/check_links.py), run it as well

    line 33
  4. Very specific:

    Tag consistency audit: Use Grep to find all tags: sections across .md files.

    line 35
  5. Very specific:

    Orphaned files: Check which files are NOT referenced in any other file's related: field.

    line 41
  6. Distinctive:

    Coverage gaps: Check each directory for content:

    line 48
  7. Distinctive:

    Look for directories with only 1-2 files (may need more content)

    line 50
  8. Distinctive:

    Fix: Ensure you're in the knowledge base root.

    line 105
  9. Very specific:

    Fix: Ensure --- delimiters are present.

    line 113

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

/audit

Audit the knowledge base for quality issues: missing frontmatter, broken links, tag inconsistencies, orphaned files, and coverage gaps. Works on any markdown-heavy project.

Steps

  1. Parse focus area from $ARGUMENTS (optional). If provided, focus on that area (e.g., "tags", "frontmatter", "links"). If empty, run full audit.
  2. Find all markdown files: Use Glob to find all .md files, excluding common non-content directories: .venv/, node_modules/, .git/, archive/, .archive_old/.
  3. Frontmatter audit: First, scan a sample of existing files (first 10-20) to detect the frontmatter schema in use (which fields exist, what values are common for type and status). Then for each markdown file, check:
    • Has YAML frontmatter (starts with --- and has closing ---)
    • Core fields present: title, tags (and any other fields consistently used across the KB)
    • type and status values (if used) are consistent with the detected schema
    • tags is a non-empty list Track files missing frontmatter and files with incomplete/invalid frontmatter.
  4. Link check: Look for broken internal links:
    • Grep for markdown links \[.*\]\(.*\.md\) and verify each target file exists
    • If a link-checking script exists in the project (e.g., scripts/check_links.py), run it as well
  5. Tag consistency audit: Use Grep to find all tags: sections across .md files. Look for:
    • Near-duplicate tags (e.g., "ai" vs "AI" vs "artificial-intelligence")
    • Tags used only once (potential typos)
    • Very common tags that might be too broad List all unique tags with counts.
  6. Orphaned files: Check which files are NOT referenced in any other file's related: field. Files that exist but are never cross-referenced may be orphaned.
  7. Content quality: Find documents that appear to be ideas or opportunities (based on detected type field or directory location) and check:
    • Documents still in draft status for more than 30 days
    • Documents missing key metadata fields that other similar documents have
    • Documents with very little content (< 100 words, excluding frontmatter)
  8. Coverage gaps: Check each directory for content:
    • Flag any empty or near-empty directories
    • Look for directories with only 1-2 files (may need more content)
  9. Output report:
    ## KB Audit Report
    
    **Date:** [today]
    
    ### Summary
    - Total .md files: X
    - With frontmatter: X (X%)
    - Without frontmatter: X
    
    ### Frontmatter Issues
    | File | Issue |
    |------|-------|
    | path | Missing field: type |
    
    ### Broken Links
    [list of broken references]
    
    ### Tag Analysis
    - Total unique tags: X
    - Single-use tags: [list]
    - Potential duplicates: [list]
    
    ### Orphaned Files
    [files not referenced anywhere]
    
    ### Content Quality
    - Stale drafts (> 30 days): [list]
    - Missing metadata: [list]
    - Low-content files: [list]
    
    ### Coverage
    [directory analysis]
    
    ### Recommendations
    1. [specific action]
    2. [specific action]
    3. [specific action]

Gotchas

  1. Single-use tags are not always typos — a tag used once might be intentional (e.g., event-2026 for a specific event). Flag but don't auto-merge without checking context.
  2. Orphaned files may be intentional archives — files in 3-inbox/ or 4-opportunities/ are often standalone by design. Only flag orphans in methodology or principles directories.
  3. Frontmatter schemas drift across directories — 0-principles/ uses type: principle, 3-inbox/ uses type: capture. Detect per-directory norms, don't enforce one schema globally.
  4. related: field is not authoritative — many valid cross-references exist as inline markdown links [text](path.md), not just YAML related: arrays. Check both.
  5. Empty directories may be placeholders — 5-active-projects/ or 6-companies/ might be empty by design (populated by automation). Check registry.yaml or Makefile before flagging.

Common Issues

No markdown files found

Cause: Running in wrong directory or all files excluded. Fix: Ensure you're in the knowledge base root. Check exclude patterns in step 2.

Too many single-use tags

Cause: Inconsistent tagging across documents. Fix: Pick canonical tags from the most-used list. Run audit again after cleanup.

Frontmatter validation errors

Cause: YAML syntax issues (missing quotes, wrong indentation). Fix: Ensure --- delimiters are present. Use type: and status: values consistent with your KB's detected schema.

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.