#5555th of 87 in Coding agents and assistants
code-graph-rag
Tree-sitter code knowledge graph you query and edit in plain English
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 5.3k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
2,219 on 25 Mar 20265,259 on 14 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Parses a multi-language repository with Tree-sitter into a knowledge graph stored in an embedded SQLite file by default, with Memgraph and Neo4j as opt-in backends. A CLI agent (cgr) turns natural-language questions into Cypher, retrieves source, finds dead code, and applies AST-based edits with a diff preview. It also runs as an MCP server for Claude Code and other clients.
Who it is for: Engineers wanting graph-grounded code Q&A and edits across mixed-language repos
Strengths
- Embedded SQLite graph by default: no database server, Docker or cmake needed
- 14 languages fully supported under one graph schema, plus an ast-grep tier for more
- Optional compiler frontends (libclang, go/types, Roslyn, javac, Jedi) and runtime trace merging
- Runs as an MCP server for Claude Code and other MCP clients
Weaknesses
- PyPI and GitHub releases follow every 50th version; newest code needs a git install
- Needs Python 3.12+ and ripgrep; older system interpreters need manual pinning
- --clean deletes every project in the shared graph, not only the current repo
- Managed cloud and air-gapped deployments are sold as paid enterprise services
What it needs
- Docker
- Needs Python 3.12+, ripgrep, Memgraph (optional), Neo4j (optional), Qdrant (optional)
Also in Coding agents and assistants
See all 87| Rank | Project | Score |
|---|---|---|
| 1 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 2 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 82 out of 100 |
| 3 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |
| #4 | Open SWELangChain coding agent that plans, implements and reviews pull requests | 80 out of 100 |
| #5 | opencodeTerminal coding agent with build and plan modes, plus a desktop app | 79 out of 100 |