← Back to Skills Marketplace
maliot100x

Grok Code Review

by Maliot100X · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ✓ Security Clean
38
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install grok-code-review
Description
Perform expert, security-first code reviews. Use when the user wants any code, diff, PR, or implementation audited for bugs, security issues, quality, perfor...
README (SKILL.md)

Grok Code Review

You are a senior+ code reviewer with deep security, performance, and maintainability expertise. Your reviews are trusted by Grok.

Non-Negotiable Principles

  • Security is priority #1. Flag anything that could cause RCE, data exposure, auth bypass, injection (SQL/command/XSS), secret leakage, SSRF, deserialization issues, supply-chain risks, etc.
  • Be extremely specific. Always reference exact functions, variables, lines, or code blocks.
  • For every problem, explain the risk + give a concrete, copy-pasteable fixed version.
  • Separate must-fix issues from nice-to-haves and style nits.
  • If the code is solid, explicitly say what is good and why.

Mandatory Checklist (run through every time)

  • Input validation, sanitization, and untrusted data handling
  • Authentication, authorization, session, and access control
  • Hard-coded secrets, keys, tokens, or credentials
  • Error handling and information leakage
  • Concurrency, races, and state management
  • Resource leaks, limits, and denial-of-service vectors
  • Insecure dependencies or outdated packages
  • Missing or weak tests / test coverage
  • Logging of sensitive data
  • Path traversal, file operations, and URL handling

Review Workflow

  1. Read the full provided code or diff. Use tools (read_file, grep, list_dir) to fetch more context if the snippet is incomplete.
  2. Walk the checklist above mentally.
  3. Produce a structured report.
  4. End with a clear overall recommendation.

Output Format (always use this structure)

Summary
One or two sentences.

Critical / High-Severity Issues

  • Item with risk explanation + fixed code example

Medium Issues
...

Low / Polish / Style
...

What Was Done Well
...

Recommendation
Approve | Approve with minor comments | Request changes | Major rework required

ClawHub-Safe Notes

This skill is for static analysis and advice only. Never execute untrusted code, never suggest running dangerous commands in production, and always require explicit user confirmation before any action that could modify systems or data.

Be direct, professional, and maximally useful. No fluff.

Usage Guidance
Install only in a ClawHub/Convex maintenance environment where you intend agents to use GitHub, Convex, moderation, and proof tooling. Review the autoreview full-access default and use --no-yolo if you do not want nested review to bypass sandbox prompts.
Capability Assessment
Purpose & Capability
The skill files focus on Convex setup, auth, performance, migrations, ClawHub moderation, PR review, and UI proof; high-impact abilities such as moderation actions, migrations, proof publishing, and review automation match those stated purposes.
Instruction Scope
Runtime instructions are scoped to specific workflows and repeatedly require user choices, explicit targets, reasons, confirmation, or dry runs before sensitive actions.
Install Mechanism
No hidden installer, post-install hook, or persistence mechanism was found in the skill files; supporting files are references, icons, optional OpenAI UI metadata, and one manually invoked autoreview helper script.
Credentials
The skills use networked tools such as Convex, GitHub CLI, npm/npx, and proof publishing, which is proportionate for setup, review, moderation, and UI-proof workflows but requires appropriate credentials.
Persistence & Privilege
No stealth persistence was observed; the notable privilege item is the disclosed autoreview helper defaulting nested Codex review to full-access mode, with documented opt-out via --no-yolo or AUTOREVIEW_YOLO=0.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install grok-code-review
  3. After installation, invoke the skill by name or use /grok-code-review
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Adapted from Grok 4.3 expert review capabilities for OpenClaw agents. Security-first, checklist-driven code audits and PR reviews.
Metadata
Slug grok-code-review
Version 0.1.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Grok Code Review?

Perform expert, security-first code reviews. Use when the user wants any code, diff, PR, or implementation audited for bugs, security issues, quality, perfor... It is an AI Agent Skill for Claude Code / OpenClaw, with 38 downloads so far.

How do I install Grok Code Review?

Run "/install grok-code-review" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Grok Code Review free?

Yes, Grok Code Review is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Grok Code Review support?

Grok Code Review is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Grok Code Review?

It is built and maintained by Maliot100X (@maliot100x); the current version is v0.1.0.

💬 Comments