An automated assistant that provides structured feedback on code quality, security, and performance for pull requests.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pr-review-assistant
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install pr-review-assistant using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is PR Review Assistant?
The PR Review Assistant is a sophisticated tool designed to streamline the code review process by providing automated, high-quality feedback. As a core component of the Openclaw Skills ecosystem, it evaluates code changes across six critical dimensions, ensuring that every pull request meets your project's standards for correctness, security, and maintainability. It acts as a tireless first-responder for code reviews, allowing human reviewers to focus on high-level architectural decisions while the AI handles the repetitive task of spotting edge cases and common vulnerabilities.
By integrating this assistant into your workflow, you ensure that security risks, performance bottlenecks, and technical debt are identified early in the development lifecycle. The skill leverages context-aware analysis to provide deep insights that go beyond simple linting, making it an essential asset for teams using Openclaw Skills to maintain high-velocity development without sacrificing quality.
PR Review Assistant Use Cases
- Automating the initial review pass for all incoming pull requests to reduce manual effort.
- Preparing code for peer review by identifying and fixing common issues locally using staged changes.
- Conducting targeted security audits of sensitive modules using focus-specific parameters.
- Enforcing repository-wide naming conventions and best practices automatically.
- Streamlining CI/CD pipelines by integrating automated quality gates.
How PR Review Assistant Works
- The assistant identifies the source of code changes, whether from a specific GitHub PR, staged files, or a branch diff.
- It analyzes the code across six key areas: Correctness, Security, Performance, Maintainability, Testing, and Best Practices.
- The AI agent processes the diff while considering the broader project context to understand how changes affect existing logic.
- A structured report is generated, categorizing findings into severity levels such as Must Fix, Should Fix, or Consider.
- The tool outputs the review in the desired format, such as Markdown or JSON, which can be posted directly to GitHub as a comment.
PR Review Assistant Setup
To get started with the PR Review Assistant within your Openclaw Skills setup, ensure you have your environment configured with the necessary scripts. You can run the assistant using the following CLI commands:
# Review current branch changes against main
python3 scripts/pr_review.py
# Review a specific PR (requires gh CLI)
python3 scripts/pr_review.py --pr 42
# Review staged changes only
python3 scripts/pr_review.py --staged
PR Review Assistant Data Schema & Taxonomy
The skill produces a structured evaluation report. The data is organized as follows:
| Field | Description |
|---|---|
| Risk Level | Visual indicator (Low, Medium, High) of the overall PR risk. |
| Must Fix | Blocking issues like security vulnerabilities or logic errors. |
| Should Fix | Non-blocking improvements for better performance or clarity. |
| Consider | Optional suggestions for alternative architectural approaches. |
| Positive Notes | Highlights of well-implemented logic and patterns. |
PR Review Assistant Advanced Features
- Context-Aware Analysis: Goes beyond the diff to understand surrounding code and project structure.
- Cross-File Pattern Detection: Identifies logical inconsistencies that span multiple modified files.
- Actionable Fix Suggestions: Generates specific code snippets to resolve identified issues automatically.
- Customizable Focus: Allows users to prioritize specific review areas like security or performance via flags.
- Repository Learning: Adapts to specific repository conventions and preferred coding styles over time.
SKILL.md
Loading
Related Openclaw Skills
N+1 Query Detector
A comprehensive diagnostic tool to find and fix performance-draining N+1 query patterns in ORM-based applications.
Repository Health Score
A comprehensive auditing tool that generates a 0-100 health score for any repository based on eight critical engineering dimensions.
Systemd Unit Generator
A powerful tool to generate, validate, and lint hardened systemd unit files for services, timers, and sockets.
Cluster Upgrade Planner
A systematic tool for planning Kubernetes cluster upgrades by identifying API deprecations and generating risk-aware migration runbooks.
Package JSON Linter
A comprehensive auditing tool for Node.js package.json files to ensure security, valid semver, and project best practices.
Pyproject TOML Validator
A comprehensive validation tool for Python pyproject.toml files ensuring PEP standard compliance and tool configuration integrity.
Featured*
