An enterprise-grade automated code review agent that uses the SPEAR framework to evaluate security, performance, and reliability across any programming language.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install afrexai-code-reviewer
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 afrexai-code-reviewer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Enterprise Code Review Engine is a sophisticated AI agent designed to automate the heavy lifting of peer reviews. By integrating this into your Openclaw Skills library, you gain access to the SPEAR framework—a rigorous evaluation system covering Security, Performance, Error Handling, Architecture, and Reliability.
It acts as a tireless senior developer, scanning PRs, local diffs, and specific files for vulnerabilities and architectural smells without requiring any external dependencies. This skill ensures consistency across your codebase, catching critical issues like SQL injection, N+1 queries, and unhandled promises before they reach production. It is an essential addition for teams looking to maintain high standards through Openclaw Skills.
To get started with this skill in your Openclaw Skills environment, ensure you have the GitHub CLI installed for remote reviews. Use the following commands to trigger the engine:
# Review a specific GitHub Pull Request
gh pr diff 42 --repo owner/repo | [agent-trigger] "Review this PR"
# Review local staged changes before committing
git diff --cached | [agent-trigger] "Review these changes"
# Review a specific file for security issues
[agent-trigger] "Review src/auth/login.ts for security vulnerabilities"
The skill organizes its analysis into a structured format to ensure clarity within the Openclaw Skills ecosystem:
| Component | Description |
|---|---|
| SPEAR Scoreboard | A summary table scoring each dimension from 1-10 with key findings. |
| Finding Severity | Classified as CRITICAL (-3pts), HIGH (-2pts), MEDIUM (-1pt), or LOW (-0.5pts). |
| Verdict Thresholds | Final scores from 0-100 (e.g., 90-100 is EXCELLENT; 0-39 is BLOCK). |
| Fix Snippets | Markdown code blocks providing the exact refactored code required to resolve a finding. |
Loading
A specialized AI agent skill that analyzes cloud billing data to identify waste, calculate ROI, and provide a prioritized 90-day savings roadmap.

A comprehensive retention operating system that uses automated scoring models and playbooks to transform clients into long-term revenue engines.

A comprehensive operational agent for managing and scaling commercial and residential cleaning businesses through structured data and workflows.

A complete methodology for maximizing developer productivity and shipping production-grade code 10X faster using Claude Code.

A comprehensive 12-phase system designed to build, engage, and monetize thriving communities across platforms like Discord, Slack, and Telegram.

A strategic framework for building data-driven compensation structures, salary bands, and equity packages to attract and retain top talent.








































