jcodemunch-mcp-code-retrieval

Index GitHub repositories with tree-sitter AST parsing and retrieve exact code symbol implementations.

7|2|Updated May 16, 2026
One-click install
npx skills add https://github.com/Aradotso/mcp-skills --skill jcodemunch-mcp-code-retrieval
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: jcodemunch-mcp-code-retrieval
Source: https://github.com/Aradotso/mcp-skills/tree/main/skills/jcodemunch-mcp-code-retrieval
Command: npx skills add https://github.com/Aradotso/mcp-skills --skill jcodemunch-mcp-code-retrieval

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sentence-transformers, torch.

What problem does it solve?

It solves the problem of wasting tokens and time when exploring large codebases by repeatedly scanning whole files to find specific functions, classes, or constants.

Core Features & Use Cases

  • Structured code indexing with tree-sitter: Builds an AST-based index once so symbol lookups are fast and precise.
  • Token-efficient symbol retrieval: Returns exact implementations (rather than broad file snippets), targeting major token savings for code reading.
  • Context assembly under a token budget: Orchestrates the right tool sequence to produce ranked, task-relevant context for bug fixing, refactoring, or understanding unfamiliar systems.

Example: An engineer needs to fix an auth-related bug but doesn’t know where the correct implementation lives; they can search by symbol name, fetch the exact function body, then retrieve references and blast radius before making changes.

Quick Start

Ask your AI client to index your repository and then return the exact implementation of the symbol you care about within a specified token budget.

Frequently Asked Questions about jcodemunch-mcp-code-retrieval

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

FAQPage Schema
How do I retrieve exact function implementations from a large codebase without wasting tokens?

You can index a repository using tree-sitter AST parsing to build a structural map, then query exact symbol implementations to retrieve targeted code context instead of scanning whole files.

What is tree-sitter codebase indexing and how does it help with code navigation?

Tree-sitter codebase indexing builds an AST-based map of your source code once, allowing precise symbol lookups, implementation searches, and reference tracing across common programming languages without repeatedly scanning files.

How do I assemble task-relevant context for debugging under a strict token budget?

You assemble task-relevant context by querying indexed symbols, fetching exact implementations, and retrieving references, applying a token budget constraint to rank and return only the most relevant code.

Can I use this symbol search approach with any programming language?

Symbol search applies to common programming languages through tree-sitter AST parsing, which supports multiple language grammars for building structural code indexes and performing implementation lookups.

Do I need an MCP client to perform structured symbol queries on my codebase?

Yes, an MCP client workflow is required to support repository indexing, structured symbol queries, and ranked context retrieval constrained by a token budget for effective codebase exploration.

What is the best way to trace code references and blast radius before refactoring?

The best way to trace references and blast radius is to search by symbol name, fetch the exact function body, then retrieve references using tree-sitter indexing to understand the impact of code changes.