codegraph vs code-review-graph
Local code graphs for coding agents and AI reviews
codegraph and code-review-graph both build local SQLite graphs of repository structure and expose code context through MCP and a CLI. codegraph focuses on agent-driven code exploration and keeping an index current, while code-review-graph emphasizes review context, impact analysis, and pull request workflows.
codegraph: Give Coding Agents a Local Code Knowledge Graph
CodeGraph builds and maintains a local graph of code structure so AI coding agents can retrieve relevant source, call paths, and change impact without exploring files one by one. It connects to multiple agents through MCP and supports many languages.
code-review-graph: Map Code for AI Reviews
code-review-graph builds a local, persistent map of code structure and serves targeted review context to AI coding tools through MCP and a CLI. It suits teams working in large or interconnected repositories who want impact analysis without repeatedly scanning the whole codebase.
| codegraph | code-review-graph | |
|---|---|---|
| Language | C | Python |
| License | MIT | MIT |
| Stars | 73.1k | 31.9k |
| Forks | 4.7k | 2.9k |
| Last analyzed | Oct 3, 2026 | Oct 3, 2026 |
Key differences
- codegraph offers source retrieval, symbol paths, and affected-test discovery; code-review-graph adds review context, dead-code inspection, architecture overviews, and visualization exports.
- codegraph watches files and incrementally syncs changes; code-review-graph supports incremental updates through hooks and watch mode.
- code-review-graph includes a GitHub Action for risk-scored pull request comments; codegraph describes agent configuration for Claude Code, Cursor, Codex, Gemini, and GitHub Copilot.
- codegraph is written in C and code-review-graph is written in Python; code-review-graph requires Python 3.10 or newer, while embedding codegraph's library requires Node.js 22.5 or later.
- Both use the MIT license and store graph data locally by default. code-review-graph can optionally use external embedding providers, while codegraph reports anonymous usage telemetry by default with opt-out options.
- The supplied project data lists 73.1k stars for codegraph and 31.9k for code-review-graph.
Choose codegraph if you…
- want agents to retrieve relevant source, call paths, and potential change impact while exploring code.
- need file watching and incremental index syncing for a project-specific code graph.
- want CLI commands for tracing callers and callees or finding tests related to changed files.
Choose code-review-graph if you…
- need focused AI review context and impact analysis across callers, dependents, and tests.
- want a GitHub Action for risk-scored pull request review comments.
- need CLI tools for dead-code discovery or exporting graph visualizations.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.