skills.rest
TomasGC
TomasGCCOMMUNITY
Explore all skills by this author →
GitHub

index-documentation

Discovers and categorizes project documentation files into FUNCTIONAL, TECHNICAL, and PERSONAL sections in CLAUDE.md.

Updated May 11, 2026
Search Tags:#documentation#markdown#indexing#claude.md#project context#categorization#file discovery
One-click install
npx skills add https://github.com/TomasGC/Meerkat --skill index-documentation-tomasgc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: index-documentation
Source: https://github.com/TomasGC/Meerkat/tree/main/skills/index-documentation
Command: npx skills add https://github.com/TomasGC/Meerkat --skill index-documentation-tomasgc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Projects accumulate scattered Markdown and text documentation that is hard to track, and manually maintaining an up-to-date documentation index in CLAUDE.md is tedious and error-prone. ## Core Features & Use Cases - Automatic Discovery: Recursively scans the project for all .md and .txt files while excluding build artifacts, dependencies, and temp directories. - Three-Category Organization: Classifies files into FUNCTIONAL (user-facing, requirements), TECHNICAL (architecture, API, design), and PERSONAL (*.local.md files) sections. - CLAUDE.md Integration: Updates section 4 of the Session Startup block in the project CLAUDE.md while preserving surrounding structure. - Use Case: After adding several new design docs and API references to a repository, run the skill to regenerate a clean, categorized documentation index so every session starts with full context. ## Quick Start Ask the assistant to scan the project documentation and update the documentation index in CLAUDE.md.

Frequently Asked Questions about index-documentation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automatically index all Markdown files in a project?▼

Run the index-documentation skill, which recursively globs for .md and .txt files, categorizes them into FUNCTIONAL, TECHNICAL, and PERSONAL groups, and writes the organized list into section 4 of your project CLAUDE.md.

How are documentation files categorized into functional and technical?▼

Categorization uses filename, location, and content patterns: README, user guides, and specifications become FUNCTIONAL; architecture, API, and design docs become TECHNICAL; all *.local.md files become PERSONAL. Ambiguous files trigger a clarifying question.

Which directories are excluded when scanning documentation?▼

The scan excludes node_modules, vendor, .git, bin, obj, dist, build, .tmp, and tmpclaude directories. All other .md and .txt files, including those under .claude/, are included in the index.

What happens if CLAUDE.md does not exist in the project?▼

The skill reports that CLAUDE.md with a Session Startup section was not found and offers options: create the file with a project setup flow, create it manually, or skip documentation indexing.

Are personal .local.md files kept separate from shared documentation?▼

Yes, every *.local.md file is placed exclusively in the PERSONAL section, which is clearly marked as not committed. Personal notes never appear in the FUNCTIONAL or TECHNICAL sections.

Related Skills

Explore More Software Engineering Skills→
adamrdrewadamrdrew
1

db-verify

Verify database directory existence and index file structure.

Community
Intermediate
sachio222sachio222

context-engineering

Load project context from index files before starting work.

Community
Advanced
geekdesign27geekdesign27

bmad-index-docs

Generate an index.md file listing documents with content-based descriptions.

Community
Basic
beelabstudiobeelabstudio

documentation-writer

Creates and maintains technical documentation, API docs, README files, and user guides.

Official
Basic