oracle

Bundle prompts and files for second-model API or browser review.

2|Updated Dec 8, 2018
One-click install
npx skills add https://github.com/indykish/dotfiles --skill oracle-indykish
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/indykish/dotfiles/tree/main/skills/oracle
Command: npx skills add https://github.com/indykish/dotfiles --skill oracle-indykish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use the Oracle CLI to bundle a prompt plus the right files and obtain a second-model review (API or browser) for debugging, refactors, design checks, or cross-validation.

Core Features & Use Cases

  • Bundle a prompt with a curated file set into a single request for a contextual second-model review.
  • Support API and browser modes, as well as multi-model cross-checks to validate designs and changes.
  • Use Case: verify a complex refactor by attaching related code samples and obtaining thorough feedback from models like Claude Sonnet or GPT variants.

Quick Start

Run oracle to bundle your prompt with the selected files and review the second-model feedback for debugging, refactors, and cross-validation.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I get a second AI model to review my code refactor with full context?

You can obtain a second-model review by bundling a prompt with curated code files into a single request payload. This ensures the reviewing model receives full context for debugging, refactors, and design checks.

What is the best way to cross-check code changes using multiple AI models?

The best way to cross-check changes is by using multi-model cross-checks within an API request. This validates designs and refactors by comparing feedback across multiple models like Claude Sonnet or GPT variants.

Can I use a specific AI engine for second-model code review in API mode?

Yes, second-model code review operates in API mode with explicit engine and model selections. You specify the exact model during the request to ensure the output meets your cross-validation requirements.

How do I bundle a prompt with specific files for a contextual AI review?

Bundling a prompt with files creates a single payload that carries your instructions and relevant code context together. This ensures the second-model review has direct access to necessary design and debugging context.

Does second-model code review work in a browser or do I need an API?

Second-model code review supports both API and browser modes. API mode enables explicit engine selections and automated multi-model cross-checks, while browser mode allows manual contextual review.

What guardrails are enforced when sending code files to a second model for review?

The second-model review process enforces versioning, model guidance, and safety prompts to ensure proper usage. These guardrails protect the code review payload and validate outputs during multi-model cross-checks.