claude-review

Review plans, diffs, codebases, and completed work via the local Claude CLI.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kamelmh/opencode-config --skill claude-review-kamelmh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-review
Source: https://github.com/kamelmh/opencode-config/tree/main/skills/claude-review
Command: npx skills add https://github.com/kamelmh/opencode-config --skill claude-review-kamelmh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly get a trustworthy second opinion from a local Claude CLI when you want to confirm a plan, review code changes, validate completed work, or check an exploit chain without writing a custom prompt yourself.

Core Features & Use Cases

  • Plan confirmation: Sanity-check implementation plans before you start building.
  • Code review: Review diffs or an entire codebase for bugs, quality issues, and design risks.
  • Double-checking: Verify that completed work actually matches the original task and does not miss edge cases.
  • CTF and exploit-chain review: Validate whether each step in an attack chain is logically feasible under the stated constraints.
  • Result summarization: Returns findings in a structured format that separates confirmations, real issues, nits, and recommendations.

Quick Start

Ask this skill to have Claude review your plan, diff, full codebase, or exploit chain and summarize the findings clearly.

Frequently Asked Questions about claude-review

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

FAQPage Schema
How do I get an automated code review on a diff using Claude CLI?

Automated code review on a diff uses Claude CLI to parse changes non-interactively and surface actionable findings like bugs, quality issues, and design risks.

Can I use Claude to validate CTF exploit-chain steps before execution?

Yes, Claude can validate CTF exploit-chain steps by checking whether each attack chain step is logically feasible under stated constraints when a second opinion is explicitly requested.

What is the best way to double-check completed work against original task requirements?

The best way to double-check completed work is to request a second opinion from Claude CLI, which verifies results match the original task and identifies missed edge cases.

Do I need to write a custom prompt for Claude to review my implementation plan?

No, you do not need to write a custom prompt for plan confirmation; the Skill handles prompt construction, task-specific context gathering, and effort selection automatically.

How does code review handle summarizing findings from a full codebase analysis?

Code review summarizes full codebase analysis findings by returning results in a structured format separating confirmations, real issues, nits, and recommendations.