What is required to use this skill?

You need the GrepAI CLI installed and the MCP server running via the 'grepai mcp-serve' command to allow Claude to communicate with your local code index.

What is the primary benefit of using GrepAI with Claude?

GrepAI allows Claude to perform semantic searches and call graph analysis, meaning Claude can find code based on meaning rather than just exact text matches, and understand how different parts of your code interact.

Can I trace dependencies across my entire project?

Yes, using tools like 'grepai_trace_graph', you can build a complete call graph starting from a root symbol to a specified depth to see exactly how your functions are interconnected.

How does compact mode improve performance?

Compact mode uses abbreviated JSON keys to reduce the number of tokens in the response by approximately 80%. This saves costs and prevents Claude's context window from filling up too quickly during deep code analysis.

Does GrepAI support semantic search?

Yes, the 'grepai_search' tool uses embeddings (via providers like Ollama) to find code snippets that match the conceptual meaning of your search query.

GrepAI MCP Tools Integration

Name: GrepAI MCP Tools Integration
Author: yoanbernabeu

byyoanbernabeu

•

开发者工具

Integrates advanced semantic code search and call graph analysis into Claude using the Model Context Protocol.

This skill provides a comprehensive interface for GrepAI's Model Context Protocol (MCP) server, enabling Claude to perform deep semantic searches and complex code navigation within local repositories. It allows developers to trace function callers and callees, visualize recursive call graphs, and search codebases based on intent rather than just keywords. By leveraging features like a specialized compact mode for token efficiency and real-time index status checks, it transforms Claude into a powerful code explorer capable of understanding intricate relationships and dependencies within large-scale software projects.

主要功能

01Semantic code search using vector embeddings for intent-based discovery

02Comprehensive call graph generation to visualize function hierarchies and recursion

03Bidirectional function tracing to identify all callers and callees of a specific symbol

04Real-time indexing status monitoring to ensure search data health and consistency

05Token-efficient compact mode reducing response size by up to 80% for optimized AI context

069 GitHub stars

使用场景

01Performing impact analysis by tracing all functions affected by a proposed symbol or logic change

02Navigating large, unfamiliar codebases to find specific logic, middleware, or implementations

03Debugging complex execution flows through deep recursive call graph exploration

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yoanbernabeu/grepai-skills integration

For use in Claude.ai and ChatGPT

Download Skill