code-review

Review code changes and pull request diffs for bugs, security vulnerabilities, and quality issues.

13|1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/danchamorro/pi-agent-toolkit --skill code-review-danchamorro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/danchamorro/pi-agent-toolkit/tree/main/dotfiles/global-skills/code-review
Command: npx skills add https://github.com/danchamorro/pi-agent-toolkit --skill code-review-danchamorro

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI-powered code reviews reduce the time and cognitive load required to find bugs, security vulnerabilities, and quality regressions in code changes, enabling faster, safer iteration and more reliable pull request feedback.

Core Features & Use Cases

  • Finds bugs, security issues, and quality risks in changed code and groups findings by severity (Critical, Warning, Info).
  • Supports reviewing staged, committed, or all changes and comparing against a base branch or specific commit range.
  • Offers output modes optimized for humans or agents (detailed fixes via plain mode or minimal, AI-optimized prompts via prompt-only).
  • Enables an autonomous fix-and-verify workflow: implement changes, run reviews, create task lists, apply fixes, and re-run until acceptable quality is reached.

Quick Start

Run a CodeRabbit review on the current changes in prompt-only mode to receive concise AI-optimized findings.

Frequently Asked Questions about code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate security-focused code reviews for pull requests?

Automated AI code reviews identify bugs, security vulnerabilities, and quality issues in pull request diffs by applying static analysis to changed code, grouping findings by severity like Critical or Warning.

Can I review staged changes or specific commit ranges instead of a full repository?

Yes, code reviews support analyzing staged changes, committed changes, or full repository updates, and can compare against a base branch or a specific commit range.

Do I need the CodeRabbit CLI to run automated pull request reviews?

Yes, performing AI-powered code reviews requires the CodeRabbit CLI with authentication, enforcing secret checks and narrow token scopes before sending diffs for analysis.

What is the best way to output code review findings for other agents or humans?

Code reviews offer output modes optimized for humans or agents: plain mode provides detailed fixes for humans, while prompt-only mode delivers minimal, AI-optimized prompts for agents.

How do I run an autonomous fix-and-verify workflow for bug finding?

An autonomous fix-and-verify workflow implements changes, runs reviews, creates task lists, applies fixes, and re-runs static analysis until acceptable code quality is reached.