Codebaxing

Provides semantic code search capabilities using natural language queries, indexing codebases with Tree-sitter parsing, all-MiniLM-L6-v2 embeddings, and ChromaDB vector storage.

Codebaxing enables semantic code search across local codebases using natural language queries instead of exact text matching. It indexes source code using Tree-sitter for parsing, generates embeddings with the all-MiniLM-L6-v2 model, and stores vectors in ChromaDB for similarity search. The server supports 40+ programming languages and provides tools for searching code, viewing statistics, and managing project memories. Installation is streamlined with built-in configuration commands for Claude Desktop, Cursor, Windsurf, and other MCP-compatible editors.