code-graph-rag

About

An accurate Retrieval-Augmented Generation (RAG) system that analyzes multi-language codebases using Tree-sitter, builds comprehensive knowledge graphs, and enables natural language querying of codebase structure and relationships as well as editing capabilities.

cgr is published to PyPI and can be installed system-wide so it works from any target repo without activating a project virtualenv. Install with the treesitter-full (all languages) and semantic (vector search) extras:

For a Python-only install, omit the extras. For local development from a clone, use uv tool install --editable "/path/to/code-graph-rag[treesitter-full,semantic]".

After install, cgr is on PATH. From any…