skills.rest
goncalossilva
goncalossilvaCOMMUNITY
Explore all skills by this author →
𝕏@goncalossilvaGitHub

oracle

Send curated files to an opposite AI model family for code review.

24|3|Updated Jan 4, 2026
Search Tags:#code review#debugging#refactoring#llm#cross-model#ai review
One-click install
npx skills add https://github.com/goncalossilva/.agents --skill oracle-goncalossilva
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/goncalossilva/.agents/tree/main/skills/oracle
Command: npx skills add https://github.com/goncalossilva/.agents --skill oracle-goncalossilva

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a way to get a "second opinion" on your code or design by leveraging a different AI model family, helping to catch errors or identify alternative solutions you might have missed.

Core Features & Use Cases

  • Cross-Model Review: Get feedback from an AI model different from your current one (e.g., OpenAI models reviewing Claude, and vice-versa).
  • Targeted Analysis: Ask for specific types of feedback like debugging, refactoring advice, design checks, or code reviews.
  • Use Case: You've written a complex piece of code using a Claude model and want to ensure its robustness. You can use the Oracle skill to have a GPT model review it for potential bugs or suggest improvements.

Quick Start

Ask the oracle skill to review the files in the src directory for potential bugs and suggest refactors.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I get a cross-model AI code review for debugging and refactoring?▼

To get a cross-model AI code review, you direct a prompt with a curated set of files to an LLM from an opposite model family, such as OpenAI models reviewing Claude, to catch errors or identify alternative solutions.

What is cross-model AI review and when do I need it for my code?▼

Cross-model AI review is a technique that provides a second opinion on your code or design by leveraging a different AI model family, helping to catch errors or identify alternative solutions you might have missed.

Do I need specific CLI tools to get an AI second opinion on my code?▼

Yes, you need specific CLI tools for interacting with different AI providers to use this cross-model review functionality, as the process intelligently selects and bundles files to include in the context.

Can I use cross-model review for design checks and code reviews?▼

Yes, you can request targeted analysis for specific types of feedback including debugging, refactoring advice, design checks, and code reviews by bundling a prompt with a curated set of files.

What's the best way to review files in my src directory for potential bugs?▼

The best way to review files for potential bugs is to ask the oracle skill to review the files in the src directory, which will bundle a prompt with a curated set of files and direct the query to an opposite model family.

Why does cross-model AI review select files intelligently for the context?▼

Cross-model AI review intelligently selects files to include in the context to ensure the targeted LLM receives a curated set of relevant files, optimizing the feedback for debugging, refactoring, and code review tasks.

Related Skills

Explore More Software Engineering Skills→
edfentonedfenton
3

mern-add-auth

Configure NextAuth.js authentication with OAuth providers for MERN apps.

Community
Advanced
jpoleyjpoley
50

patch-engineer

Validate security patch quality and regression risk for code changes.

Community
Advanced
gkganesh12gkganesh12
1

gk-review

Audit source code against documentation and test suites for architectural drift.

Community
Advanced
kivo360kivo360
1

moai-cc-agents

Create and manage Claude Code sub-agents with structured agent files.

Community
Advanced