Cheatcode_
Feedback

Catalog / Engineering / Review graph

Agent skills · 7 skills

Review graph

Listed by its author as code-review-graph

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…

DoesCode reviewCoding and refactoringDebuggingGit and GitHubMemory and knowledgeArchitecture and system designMaps, travel and location

Source on GitHub ↗
curl -fsSL https://api.cheatcodehq.com/install.sh | sh -s -- code-review-graph --to all

Copies 7 skills into each app's skills folder (Claude Code: ~/.claude/skills), after checking every file against tirth8205/code-review-graph at commit 6b12d11. Nothing else on your computer changes. To undo it, run the terminal command with --uninstall at the end; that also removes what the app added.

Then start a new session in your AI app and ask for the work in plain words; it picks the skill from its description. In Claude Code you can also type /code-review-graph:build-graph.

See it in use

One real session, replayed step by step: the request, each step at the moment it happened, and the answer, unedited. Long waits are shown faster, with the real time in the corner. The two commits are invented for this example: a tiny Python project for Acme Notes, a fictional note-taking app, whose last commit changed how due dates are labeled.
The result
review.mdWhat the session produced: review.md, the file the session saved.

A saved review that names the caller the change broke, the fix, and the tests to add before pushing.

review.md, the file the session saved.

Open the whole result
Try it

After installing, start a new session and say: Use the review-delta skill to review my last commit before I push it. Tell me what it could break and which tests are missing, and save the review as review.md.

What else it can do, and its limits
  • 7 skills: build the graph, review the last commit, review a branch or pull request, debug an issue, explore a codebase and plan a safe refactor
  • Each skill calls the code-review-graph tools, which run on your computer as an MCP server or a command line (Python 3.10 or newer, installed with pip or uv)
  • Change reviews read git history, so the project must be a git repository
  • Parses many languages with tree-sitter and stores the graph in a .code-review-graph folder in the project
  • Finds callers by reading the code, not by running it: dynamic calls can be missed, as the tool itself says
  • In this recording the MCP server could not start in the sandbox, so the session used the same tool's command line

Sample input Recorded Oct 24, 2026 in Claude Code 2.1.308 with Sonnet 5.5. The video replays that session; nothing in it was acted or edited.

What's inside 7 skills

  • build-graphBuild or update the code review knowledge graph. Run this first to initialize, or let hooks keep it updated automatically.

    Build or update the code review knowledge graph. Run this first to initialize, or let hooks keep it updated automatically.

  • debug-issueSystematically debug issues using graph-powered code navigation

    Systematically debug issues using graph-powered code navigation

  • explore-codebaseNavigate and understand codebase structure using the knowledge graph

    Navigate and understand codebase structure using the knowledge graph

  • refactor-safelyPlan and execute safe refactoring using dependency analysis

    Plan and execute safe refactoring using dependency analysis

  • review-changesPerform a structured code review using change detection and impact

    Perform a structured code review using change detection and impact

  • review-deltaReview only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.

    Review only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.

  • review-prReview a PR or branch diff using the knowledge graph for full structural context. Outputs a structured review with blast-radius analysis.

    Review a PR or branch diff using the knowledge graph for full structural context. Outputs a structured review with blast-radius analysis.

About

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

Where this comes from. The public repository tirth8205/code-review-graph at commit 6b12d11, under its MIT license, copied without changes. tirth8205 is not affiliated with Cheatcode and has not claimed this listing yet. New releases are repackaged after review, never synced automatically. If this is your project, write to admin@cheatcodehq.com to claim or remove it.

Something wrong with this listing, or it did not install? Tell us

Versions

v0.0.0-g6b12d11Listed from tirth8205/code-review-graph at 6b12d11 · 2026-09-28