A powerful single-pass code review tool that identifies and automatically fixes critical bugs and security issues before your code reaches production.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pr-review
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install pr-review using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The PR Review skill is a specialized utility designed for Openclaw Skills to streamline the software development lifecycle by catching errors early. It performs deep, context-aware analysis of code changes using a single-pass review process, eliminating the overhead of multi-agent orchestration while maintaining high thoroughness. This skill is specifically built to bridge the gap between local development and formal peer review, ensuring that only high-quality, secure code is submitted to the repository.
By leveraging the capabilities of Openclaw Skills, this tool doesn't just point out problems—it actively resolves them. It evaluates code against five key dimensions: correctness, security, reliability, performance, and quality. Whether you are reviewing a small feature branch or auditing a massive legacy codebase, this skill provides actionable insights and automated patches that align with your project's specific coding standards and technical stack.
The PR Review skill requires git to be installed on your system. It is natively integrated into the environment for Openclaw Skills. You can trigger the review process using the following commands:
# Review current branch changes
/pr-review
# Audit a specific directory
/pr-review src/api/
# Run a full codebase audit
/pr-review --audit
The skill organizes its analysis using a structured scoring and classification system to ensure accuracy within Openclaw Skills:
| Attribute | Description |
|---|---|
| Mode | Diff (changed files) or Audit (specified files/patterns). |
| Classification | Issues are tagged as blocker, important, or minor based on impact. |
| Scoring | 0-100 scale; scores >= 70 trigger auto-fixes in Diff mode. |
| Exclusions | Automatically ignores node_modules, dist, .git, and build directories. |
| Context Sources | Scans .eslintrc, .prettierrc, and biome.json to match project style. |
Loading
An autonomous agent designed to identify and remediate TypeScript error handling anti-patterns through state-managed workflows.

A sophisticated financial management skill for the OpenClaw framework featuring natural language processing and automated expense categorization.

A comprehensive financial data acquisition tool for Openclaw Skills providing real-time A-share market insights, fund flows, and technical indicators.

Generate comprehensive networking guides and research intelligence for professional coffee chats and meetings.

A pre-ship risk assessment tool that analyzes codebase impact and version-specific constraints for OpenClaw pull requests.

A powerful command-line interface for managing OpenPond repositories, deployments, and AI tools without a web UI.








































