# CC Auto Review

> CC Auto Review displays a pull request diff alongside findings from Claude Code on one screen. It is for macOS developers who want automated PR reviews that explain how a problem occurs and what goes wrong.

- Type: AI tool
- Page: https://completeaitraining.com/ai-tools/cc-auto-review/
- Published: 2026-11-05
- Updated: 2026-11-06
- Author: Jeroen Erne
- Jobs: IT and development
- Topics: Coding, Developer Tools, Productivity, Apple iOS Mac
- AI tools: GitHub
- Reading time: 2 min

### About CC Auto Review

CC Auto Review is a macOS application that reviews GitHub pull requests using your local installation of Claude Code. It displays AI findings beside the diff on a single screen, lets you accept, reject, or edit each one before posting, and assigns every PR a 0-100 risk score. The app is free and open source under the MIT license, requires no API key or repository clone, and works through your existing gh CLI and Claude Code setup.

### Review

CC Auto Review brings Claude Code into the PR review workflow without adding new accounts or credentials. It runs entirely on your machine, pulling PR data through the GitHub CLI and feeding only the changed files to Claude for analysis. The interface keeps the diff and the AI's findings side by side, so you decide what feedback actually reaches the pull request.

#### Key Features

- **Inline review with edit control.** Findings appear next to the diff. You can accept, reject, or edit each one before anything is posted to GitHub.
- **0-100 risk score.** A baseline score is computed from the diff and PR metadata. Claude can adjust the score by up to ±20 points, citing specific file:line references.
- **Autopilot mode.** An optional setting clears routine reviews automatically and hands PRs above a risk threshold to you with nothing posted. The autopilot runs only while the app is open and posts from your GitHub account.
- **Scoped file access.** Only the PR's changed files go into a temporary directory. Claude runs with a restricted set of tools and reads conventions from CLAUDE.md, AGENTS.md, or REVIEW.md on the base branch, so the PR cannot rewrite its own review rules.
- **Browser-based score preview.** You can paste any public PR URL on the website to run the scoring code in your browser without installing the app.

#### Pricing and Value

CC Auto Review is free and open source, published under the MIT license. There is no paid tier, no account to create, and no API key required-the app uses your existing local Claude Code and GitHub CLI installations. The maker has not announced any future pricing changes.

#### Pros

- No clone step needed; the app works with a temporary directory of changed files.
- Review conventions live on the base branch, preventing a PR from altering its own evaluation rules.
- Risk scoring gives a quick numeric signal before you read any findings.
- Full control over posted feedback-nothing reaches GitHub without your explicit accept, reject, or edit action.
- Zero additional accounts or credentials beyond your existing gh and Claude Code setup.

#### Cons

- macOS only, with a hard dependency on Claude Code 2.1.314+ and the GitHub CLI.
- Autopilot reviews post from your GitHub account and run Claude Code automatically, which requires checking Anthropic's terms or using an API key.
- Not well suited for solo developers who review only their own work and already ask Claude for feedback during coding.

CC Auto Review fits teams or individuals who handle a steady stream of incoming PRs and want a structured review step before posting AI feedback. The risk score and autopilot aim to reduce time spent on routine checks while keeping high-risk changes in human hands. If your workflow already covers these needs through Claude's MCP integration with GitHub, the added value here comes from the side-by-side diff view, the mandatory human gate on findings, and the scoring layer that helps prioritize which PRs to examine first.

---

This is the Markdown form of https://completeaitraining.com/ai-tools/cc-auto-review/. The whole site is mapped at https://completeaitraining.com/llms.txt.