code-review-graph

code-review-graph is an open-source AI developer tool for local-first code intelligence and context reduction for AI code review. The official source is the public GitHub repository at https://github.com/tirth8205/code-review-graph. During this creation run, repository metadata showed 18,676 stars, 2,000 forks, 142 open issues, Python as the main language, and a latest public push dated 2026-06-14. The upstream description says: Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.. That makes the project relevant for builders who are already testing coding agents, code review automation, local context systems, or AI-assisted development workflows. This is not a foundation model listing and not a generic article. It is a practical developer tool with source code that teams can inspect before trusting it with private repositories. The safest evaluation path is direct. Read the README, inspect the commands it asks you to run, and test the smallest workflow inside a disposable repository. Check what files the tool reads, whether it writes persistent state, which model providers or local services it expects, and how it behaves when a repository is large, messy, or security-sensitive. That review matters because AI coding tools can touch production code paths if they are adopted too quickly. code-review-graph is most useful for developers, AI engineering teams, and platform teams that want a clearer loop around agent work. A solo builder can use it to experiment with better code context or review behavior. A team can compare it against its current coding assistant setup and decide whether it reduces repeated prompting, narrows context, or improves review quality without hiding important implementation details. Pricing is listed as free/open-source access because the source repository is public. Real operating cost may still include connected LLM APIs, local compute, hosted runners, storage, or private infrastructure. Review the MIT License terms and upstream documentation before using it commercially. If the project later adds hosted plans or formal pricing, this page should be updated with those facts rather than duplicated. Verification note: OpenTools used the repository URL, GitHub API metadata, README content, and the queue source signal as source material. The page does not infer private roadmap details, hidden benchmarks, or paid features that are not documented upstream. README excerpt reviewed from the source repository: code-review-graph Stop burning tokens. Start reviewing smarter. English | 简体中文 | 日本語 | 한국어 | हिन्दी Usage · Commands · FAQ · Troubleshooting · GitHub Action · Reproducing the benchmarks · Roadmap AI coding tools can end up re-reading large parts of your codebase on review tasks. `code-review-graph` fixes that. It builds a structural map of your code with Tree-sitter, tracks changes incrementally, and gives your AI assistant precise context via MCP so it reads only what matters. --- ## Quick Start ```bash pip install code-review-graph # or: pipx install code-review-graph code-review-graph install # auto-detects and configures all supported platforms code-review-graph build # parse your codebase ``` One command sets up everything. `install` detects which AI coding tools you have, writes the correct MCP configuration for each one, installs p