Does this skill replace Claude's primary decision-making?

No, Claude retains final judgment. The peer consultant provides alternative viewpoints which Claude then evaluates against your project's specific requirements.

What API keys do I need for this skill to work?

This skill requires an OpenAI API key (OPENAI_API_KEY) stored in your .env.local, .env, or system environment variables to communicate with the peer models.

How do I provide project context to the peer consultant?

Context is passed manually using the --files flag to include specific code files and the --message flag to describe the problem or decision you are facing.

Which models are supported by the Second Opinion skill?

It defaults to GPT-5-pro but supports gpt-5, gpt-5-mini, and gpt-5-nano, which can be configured via the SECOND_OPINION_MODEL environment variable.

Second Opinion Peer Review

Name: Second Opinion Peer Review
Author: alberduris

byalberduris

0•

Herramientas de Colaboración

Consults external LLMs to provide alternative perspectives, technical validation, and peer reviews for complex architectural decisions.

The Second Opinion skill provides a meta-cognitive boost to Claude Code by allowing it to seek external validation from peer LLMs. Designed for high-stakes technical decision-making, this skill enables Claude to cross-reference its logic with models like GPT-5-pro to identify edge cases, explore alternative implementation patterns, and perform rigorous peer reviews. It is particularly useful when navigating complex refactors or architectural shifts where a 'fresh take' can prevent technical debt and ensure robust solution design.

Características Principales

01Support for meta-cognitive augmentation and sanity checking

020 GitHub stars

03Integration with high-reasoning models including GPT-5-pro

04Deep context passing for file-specific code reviews

05Configurable model selection via environment variables

06Synchronous execution for real-time decision support

Casos de Uso

01Gaining a fresh perspective on stubborn bugs or performance bottlenecks

02Validating complex system architectures before beginning implementation

03Performing automated peer reviews to ensure code quality and best practices

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alberduris/skills second-opinion

For use in Claude.ai and ChatGPT

Download Skill