# MCP

> Use tree-sitter-language-pack mcp reference documentation.

- Skill: `kreuzberg-dev/mcp-3` (Agent Skill)
- Install (CLI): `npx skillmds@latest add kreuzberg-dev/mcp-3`
- Raw SKILL.md: https://api.skillmd.com/api/skills/kreuzberg-dev/mcp-3/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML, Web & Frontend
- Author: kreuzberg-dev (https://skillmd.com/u/kreuzberg-dev)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/kreuzberg-dev/mcp-3

---


<!-- This file is auto-generated by alef — DO NOT EDIT. -->
<!-- alef:hash:83a7319bfcbfb7df69c7a64f73d0f721268dfe3a5dd7df3c3b0ef1d4dba614d1 -->
<!-- To regenerate: alef docs -->
<!-- To verify freshness: alef verify -->

# tree-sitter-language-pack MCP Reference

Use this skill when integrating tree-sitter-language-pack through Model Context Protocol tools, prompts, or resources.

- `cache_dir`: Return the effective parser cache directory path.
- `clean_cache`: Delete all cached parser libraries from the cache directory.
- `detect_language`: Detect the language for a file path or source content. Returns the detected language name.
- `download`: Download parser libraries from the remote registry. Pass languages list, groups, or all=true. \
                       Set fresh=true to clean the cache first.
- `info`: Show whether a language is known, downloaded, and its cache path.
- `list_languages`: List languages. source: 'available' (default), 'downloaded', or 'manifest'. Optional substring filter.
- `parse`: Parse source code with a tree-sitter grammar. Returns the syntax tree as sexp or JSON.
- `process`: Run the code-intelligence pipeline on source code. Extracts structure, imports, exports, symbols, and more.