doc-lister

Architecture: Operation-specific skill (Layer 3) </CONTEXT>

<CRITICAL_RULES>

• ALWAYS scan specified directory
• ALWAYS parse frontmatter for filtering
• NEVER modify files
• ALWAYS return structured list </CRITICAL_RULES>

Optional filters:

• doc_type - Filter by fractary_doc_type
• status - Filter by status (draft, published, deprecated)
• tags - Filter by tags (array)
• date_range - Filter by created/updated dates </INPUTS>

2. Parse Frontmatter

   • Extract YAML frontmatter from each file
   • Build document list with metadata

3. Apply Filters

   • Filter by doc_type if specified
   • Filter by status if specified
   • Filter by tags if specified
   • Filter by date range if specified

4. Return List

{
  "total": 42,
  "filtered": 15,
  "documents": [
    {
      "path": "docs/api/auth/login/README.md",
      "title": "Login Endpoint",
      "doc_type": "api",
      "status": "published",
      "version": "1.2.0",
      "updated": "2025-01-15"
    }
  ]
}