skills.rest
swtch1
swtch1COMMUNITY
Explore all skills by this author →
GitHub

second-opinion

Review code and technical decisions using Codex and Gemini models.

Updated Aug 23, 2026
Search Tags:#architecture#code review#validation#codex#gemini#ai feedback
One-click install
npx skills add https://github.com/swtch1/dotfiles --skill second-opinion-swtch1
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: second-opinion
Source: https://github.com/swtch1/dotfiles/tree/main/.claude/skills/second-opinion
Command: npx skills add https://github.com/swtch1/dotfiles --skill second-opinion-swtch1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides an external AI perspective on your code or technical decisions, helping to validate approaches and identify potential issues you might have missed.

Core Features & Use Cases

  • Codebase-Aware Review: Utilizes Codex to analyze code within the context of the entire repository.
  • Factual Verification: Employs Gemini to cross-reference technical claims and API/SDK correctness.
  • Dual-Model Analysis: Can leverage both Codex and Gemini for comprehensive review, especially for complex plans or implementations.
  • Use Case: When you've drafted a new API endpoint and want to ensure it adheres to best practices and correctly uses the SDK, you can invoke this skill for a thorough review.

Quick Start

Ask the second-opinion skill to review the proposed architecture for the new user authentication module.

Frequently Asked Questions about second-opinion

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

FAQPage Schema
How do I get external AI validation for a proposed software architecture?▼

External AI validation for software architecture is provided by cross-referencing technical claims using Codex and Gemini models. This dual-model analysis scopes plan reviews, implementation verification, and API correctness checks to identify potential issues in complex proposals.

Can I use AI code review to verify API and SDK correctness?▼

AI code review can verify API and SDK correctness by employing Gemini to cross-reference technical claims. This ensures newly drafted endpoints adhere to best practices and correctly utilize the target SDK within the full codebase context.

How does dual-model code review work for complex technical decisions?▼

Dual-model code review works by leveraging both Codex and Gemini to analyze code and technical decisions comprehensively. Codex evaluates repository context while Gemini validates factual claims, satisfying requirements for cross-model validation on complex implementations.

What is the best way to review a new authentication module implementation?▼

The best way to review a new authentication module implementation is to invoke an external AI perspective that scopes architecture and implementation verification. This validates the proposed approach and highlights potential issues missed during initial development.

Do I need to provide full repository access for codebase-aware AI feedback?▼

Codebase-aware AI feedback requires repository context for Codex to analyze code accurately. Providing this access ensures the external review properly scopes implementation verification and technical claim validation against the entire project structure.

Related Skills

Explore More Software Engineering Skills→
quanganh208quanganh208

ck:plan

Plan technical implementations and roadmaps for software projects.

Community
Advanced
2zcory2zcory

dialectical-review

Resolve contradictions in proposals and workflows to synthesize robust outcomes with residual risks.

Community
Advanced
htlin222htlin222
77

python

Generate and refactor idiomatic Python code with performance optimization.

Community
Intermediate
delorenjdelorenj

skf-create-skill

Compile and verify AI skills from YAML briefs and source code.

Community
Advanced