Skip to main content
The CodeRabbit CLI is in Beta.

​
Review local changes

This page covers local CLI code reviews. For code reviews on pull requests, see Introduction.
The CodeRabbit CLI analyzes your uncommitted changes using the same pattern recognition that powers our PR reviews. Review code as you write it, apply suggested fixes instantly, and catch critical issues while you still have full context in memory. You can get your code reviewed before you commit and get feedback immediately when the code is fresh in your mind.

​
Key features

Review uncommitted changes

Analyze your working directory for race conditions, null pointer exceptions, and logic errors before you commit.

One-step fixes

Apply simple fixes like import corrections instantly. Hand complex architectural issues to your AI agent with full context.

Context-aware reviews

Paid users get reviews enhanced by learnings from your team’s patterns - remembers your error handling styles, architectural decisions, and coding preferences.

Detects coding agent files

Automatically reads claude.md, .cursorrules, and custom team standards to enforce your specific coding guidelines in every review.

​
AI agent integration

You’re already using CodeRabbit for PR reviews. With the CLI, you can now catch issues before you commit - reviewing uncommitted changes while the code is fresh in your mind. CodeRabbit detects the problems, then your AI coding agent implements the fixes.
Claude Code users: Claude Code now supports CodeRabbit through a native plugin. See the Claude Code integration guide for the recommended plugin-based setup using /coderabbit:review instead of the CLI commands shown below.

​
Example prompt for your AI agent

Here’s a complete prompt you can use with Cursor, Codex, or other AI coding agents: 
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.

​
Components of a good prompt

Breaking down what makes an effective CodeRabbit + AI agent workflow:
1

Run CodeRabbit CLI

Tell your AI agent to run CodeRabbit with the --prompt-only flag:
coderabbit --prompt-only
You can also specify review types or branches:
# Review only uncommitted changes
coderabbit --prompt-only --type uncommitted

# With specific base branch
coderabbit --prompt-only --base develop
2

Run in the background

CodeRabbit reviews can take 7-30+ minutes depending on the scope of changes. Instruct your AI agent to run CodeRabbit in the background and set up a timer to check periodically:
Run coderabbit --prompt-only --type uncommitted in the background, let it take as long as it needs,
and check on it periodically.
3

Evaluate and implement fixes

Once CodeRabbit completes, have your AI agent evaluate the findings and prioritize:
Evaluate the fixes and considerations. Fix major issues only, or fix any critical
issues and ignore the nits.
This keeps your agent focused on meaningful improvements rather than minor style issues.
4

Verify with a second pass

Run CodeRabbit one more time to ensure fixes didn’t introduce new issues:
Once those changes are implemented, run CodeRabbit CLI one more time to make sure
we addressed all the critical issues and didn't introduce any additional bugs.
5

Set loop limits

Prevent infinite iteration by setting clear completion criteria:
Only run the loop twice. If on the second run you don't find any critical issues,
ignore the nits and you're complete. Give me a summary of everything that was
completed and why.
This ensures your AI agent completes the task efficiently and provides a clear report.

​
Integration guides

See detailed workflows for specific AI coding agents:

Claude Code integration

Automated workflow with background execution and task-based fixes

Codex integration

Integrated code review and fix implementation with Codex CLI

​
Pricing and capabilities

Free tier

Basic static analysis with limited daily usage.
  • Catches syntax errors, obvious logic issues, and common security patterns.
  • Perfect for individual developers trying out the CLI workflow.
  • Rate limit: 2 reviews per hour

Paid plans

Enhanced reviews powered by learnings from your codebase history plus higher rate limits.Paid users with linked GitHub accounts get:
  • Learnings-powered reviews: Remembers your team’s preferred patterns for error handling, state management, and architecture
  • Full contextual analysis: Understands your imports, dependencies, and project structure
  • Team standards enforcement: Applies your documented coding guidelines automatically
  • Advanced issue detection: Spots subtle race conditions, performance bottlenecks, and security vulnerabilities
  • Rate limits:
    • Free, OSS: 2 reviews per hour
    • Trial: 5 reviews per hour
    • Pro: 8 reviews per hour
Rate limits are per developer seat and subject to change.
Paid users get nearly all PR review features in the CLI, including learnings and contextual analysis. Only chat, docstrings, and unit test generation remain exclusive to PR reviews.
Contact [email protected] for custom rate limits or enterprise needs.
Platform support: Currently available on Apple (Intel and Apple Silicon) and Linux. Windows support is not available at this time.

​
Video demo

See CodeRabbit CLI in action with Claude Code:

​
Getting started

1

Install CLI

Download and install the CodeRabbit CLI to start reviewing code locally.
curl -fsSL https://cli.coderabbit.ai/install.sh | sh
2

Restart your shell

After installation, restart your shell or reload your shell configuration.
# Restart your shell or run:
source ~/.zshrc
3

Authenticate

Link your CodeRabbit account to enable personalized reviews based on your team’s patterns.
coderabbit auth login
# or use the short alias
cr auth login
Follow the browser redirect to sign in and copy the access token back to your CLI.
4

Review your code

Analyze your Git repository for issues using the CodeRabbit CLI.
coderabbit
# use plain mode or prompt-only mode when using it with Claude Code, Codex CLI, Gemini CLI, etc.
coderabbit --plain
If your main branch is not main, specify your base branch:
coderabbit --base develop
# or for other base branches
coderabbit --base master
CodeRabbit scans your working directory and provides specific feedback with suggested fixes.
5

Apply suggestions

Review findings in your terminal and either apply quick fixes or send complex issues to your AI coding agent.

​
Review modes

The CLI offers different output formats to fit your workflow: 
# Interactive mode (default) - full interface with browsable findings
coderabbit

# Plain text mode - detailed feedback with fix suggestions
coderabbit --plain

# Prompt-only mode - minimal output optimized for AI agents
coderabbit --prompt-only

​
Working with review results

CodeRabbit analyzes your code and surfaces specific issues with actionable suggestions. Each finding includes the problem location, explanation, and recommended fix. Example findings include:
  • Race condition detected: “This goroutine accesses shared state without proper locking”
  • Memory leak potential: “Stream not closed in error path - consider using defer”
  • Security vulnerability: “SQL query uses string concatenation - switch to parameterized queries”
  • Logic error: “Function returns nil without checking error condition first”

​
Browse and apply suggestions

In interactive mode, use the arrow keys to navigate to a finding and press enter to see the detailed explanation and suggested fix inline in your CLI. For simple issues like missing imports, syntax errors, or formatting problems, choose Apply suggested change to fix immediately.

​
Use AI coding agents

For AI agent integration, see the AI agent integration section for detailed workflow guidance and integration guides.

​
Managing comments

  • Ignore: Hide findings you want to address later
  • Restore: Click collapsed findings in the sidebar to show again

​
Command reference

CommandDescription
coderabbitRun code review (interactive mode by default)
coderabbit --plainOutput detailed feedback in plain text format
coderabbit --prompt-onlyShow minimal output optimized for AI agents
coderabbit authAuthentication commands
coderabbit reviewAI-driven code reviews with interactive or plain text output
crShort alias for all coderabbit commands

​
Additional options

OptionDescription
-t, --type <type>Review type: all, committed, uncommitted (default: “all”)
-c, --config <files...>Additional instructions for CodeRabbit AI (e.g., claude.md, coderabbit.yaml)
--base <branch>Base branch for comparison
--base-commit <commit>Base commit on current branch for comparison
--cwd <path>Working directory path
--no-colorDisable colored output
PR reviews and CLI reviews will differ, even if run on the same code. CLI reviews optimize for immediate feedback during active development, while PR reviews provide comprehensive team collaboration context and broader repository analysis.

​
Uninstall

Remove CodeRabbit CLI based on how you installed it.

​
If installed using install script

rm $(which coderabbit)
This removes the CodeRabbit binary from your PATH.

​
If installed using Homebrew

brew remove coderabbit