CodeRabbit started with AI-powered code reviews in pull requests. In May, we brought that same intelligence to VS Code, Cursor, and Windsurf. Now, we're extending the AI code reviews developers love into the command line with CodeRabbit CLI. In case you’re wondering, that makes us the most comprehensive AI code review tool available. We work everywhere you work.
CodeRabbit CLI helps devs perform self-reviews of code directly in their terminal. By providing automated, intelligent code analysis capabilities, it empowers developers to catch issues early, maintain consistent code standards, and make coding autonomous through seamless integration with AI coding agents in the CLI.
Vibe checking your code – now in CLI
CodeRabbit CLI delivers the same comprehensive analysis that makes our PR and IDE reviews effective at catching bugs early. CodeRabbit CLI is free to use with rate limits but with a Pro plan you can enjoy much higher limits and additional features, including:
- Context-aware analysis: Leverages your Git integration to synthesize insights from 40+ sources including static analysis tools, security scanners, and our codegraph relationship feature for the most comprehensive reviews.
- Pre-commit reviews: Analyze changes before they leave your machine for multi-layered reviews.
- One-click fixes: Apply simple fixes instantly or send complex issues to AI agents with full context hand-off.
- Coding guidelines: Auto-detects agent.md, claude.md, Cursor rules, and other coding agent configuration files.
CodeRabbit CLI: Works everywhere, with everything
Terminal-native means CodeRabbit CLI works with:
- Any Terminal App/IDE: iTerm2, Ghostty, Neovim, Lazyvim
- Any AI Coding CLI agent: Claude Code, Codex, Cursor, Gemini, OpenCode and more
How to use CodeRabbit CLI with AI Coding Agent CLI
The CodeRabbit CLI opens up new integration possibilities with AI coding agents. Here's how it works with Claude Code:
While working on a coding task, you can prompt Claude Code to use CodeRabbit and to fix any issues it finds. This is particularly useful if it’s coding from a PRD, or a tasklist.
Please implement phase 7.3 of the planning doc and then run coderabbit --prompt-only, let it run as long as it needs (run it in the background) and fix any issues.
- Claude Code will carry on the coding task and run coderabbit --prompt-only in the background. It may setup a timer interval to check on CodeRabbit. Alternatively, you can also prompt Claude to check if CodeRabbit is complete.
CodeRabbit running in the background within Claude Code
- Claude Code will then read the output of CodeRabbit which, by using the --prompt-only flag, provides the output as plain text with prompts for AI agents to read. Claude will then create a tasklist addressing each of the issues surfaced by CodeRabbit.
For Claude Code integration and automated workflows, check the CLI documentation for setup.
The CLI has two modes: interactive and plain response , making it easy to integrate into automated workflows or pass results to other tools.
Getting started
CodeRabbit CLI is available now. Install and try your first review:
#install CodeRabbit
curl -fsSL https://cli.coderabbit.ai/install.sh | sh
#Run a review in interactive mode
coderabbit
Top comments (0)