skills.rest
gideonshaked
gideonshakedCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

claude-second-opinion

Audit code for bugs, dead code, technical debt, and architectural issues.

12|Updated Apr 28, 2020
Search Tags:#architecture#security#audit#code-review#ai-audit#independent-review
One-click install
npx skills add https://github.com/gideonshaked/dotfiles --skill claude-second-opinion-gideonshaked
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-second-opinion
Source: https://github.com/gideonshaked/dotfiles/tree/main/agents/codex/skills/claude-second-opinion
Command: npx skills add https://github.com/gideonshaked/dotfiles --skill claude-second-opinion-gideonshaked

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Audit Codex work with Claude in a constrained, read-only mode to uncover bugs, dead code, technical debt, and architectural issues.

Core Features & Use Cases

  • Read-only audit: Claude reviews code and design for quality, consistency, and potential risks.
  • Focused scope: Supports path/file focus, plain-text focus, or full-repo audits.
  • Independent review: Provides an objective second opinion to guide fixes and refactors.

Quick Start

Run the claude-audit helper next to this SKILL.md with an optional focus path to perform a targeted audit.

Frequently Asked Questions about claude-second-opinion

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

FAQPage Schema
How do I perform an independent code audit on Codex work?▼

To perform an independent code audit on Codex work, use the Claude read-only helper to review code for bugs, dead code, technical debt, and architectural issues. It operates in a constrained mode to provide an objective second opinion safely.

Can I scope a code review to specific files or paths?▼

Yes, you can scope a code review to specific files or paths. The audit supports path or file focus, plain-text focus reviews, and full repository assessments depending on your review requirements.

What is a read-only AI audit and how does it work?▼

A read-only AI audit uses Claude to review code and design for quality, consistency, and potential risks without modifying the repository. It provides an independent assessment to guide fixes and refactors.

How do I control API timeout and budget limits for a code review?▼

You can control API timeout and budget limits for a code review by configuring the CLAUDE_AUDIT_TIMEOUT_SECONDS and CLAUDE_AUDIT_MAX_BUDGET_USD environment variables before running the audit helper script.

Are there limitations to using an independent AI review for technical debt?▼

The independent AI review is constrained to read-only operations, meaning it cannot directly apply fixes or refactors to your codebase. It solely identifies technical debt and architectural issues to guide your manual corrections.

Does the Claude second opinion audit work for broad repository assessments?▼

Yes, the Claude second opinion audit works for broad repository assessments. It evaluates the entire codebase for architectural issues and technical debt, while also supporting targeted path or plain-text focus reviews.

Related Skills

Explore More Software Engineering Skills→
elasticdotventureselasticdotventures
14

direnv-pattern

Automate loading of per-project environment variables via direnv and dotenv with Rust validation.

Community
Advanced
CarGDevCarGDev

Code Auditor

Audit code quality by detecting dead code, complexity, duplication, and architectural smells.

Community
Intermediate
aberrantCodeaberrantCode
1

code-review

Enforces automated code review across multiple AI engines for commits, PRs, and deployments.

Community
Advanced
IndustrialIndustrial

code-reviewer

Analyze code for security vulnerabilities and performance issues.

Community
Advanced