codesearch MCP Server

Example Use Cases

• 1

  Perform local semantic code search with minimal token usage.

• 2

  Real-time, incremental indexing and git branch-aware codebase updates.

• 3

  Privacy-first, offline AI code understanding without external API calls.

Description

codesearch provides a Model Context Protocol (MCP) server designed to bridge AI agents and codebases by enabling natural language semantic code search locally. It features token-efficient AI integration, real-time index updates, incremental indexing, and privacy-first local processing with no data leaving the machine. The server supports hybrid search methods, embedding caching, and automatic git integration for seamless and fast code understanding.