oracle

Bundle a prompt and selected files into a one-shot advisory review request.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/gmickel/claude-code-config --skill oracle-gmickel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/gmickel/claude-code-config/tree/main/skills/oracle
Command: npx skills add https://github.com/gmickel/claude-code-config --skill oracle-gmickel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Bundle a prompt plus the right files into a single request to obtain a second-model review for debugging, refactors, design checks, or cross-validation.

Core Features & Use Cases

  • One-shot review requests: combine prompt + files for a concise advisory pass from GPT-5.2 Pro, Claude, or Gemini.
  • Engine-agnostic validation: supports browser or API modes to compare design decisions across platforms.
  • File-set targeting: select files with globs or specific paths to focus the review on relevant code or docs.

Quick Start

Provide a minimal task description and run the Oracle CLI to bundle your prompt and files into a single review request.

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 for debugging and refactoring?

You can bundle a prompt and selected files into a one-shot request to a second AI model for advisory code review, debugging, refactors, design checks, or cross-validation. This approach targets relevant code by accepting a focused file set.

Can I use browser and API modes for AI code review?

Yes, AI code review supports both browser and API engine modes. This engine-agnostic validation allows you to compare design decisions and cross-validate outputs across different platforms like GPT-5.2 Pro, Claude, or Gemini.

How do I target specific files for an AI review request?

You can target specific files for an AI review request by using globs or explicit file paths. This file-set targeting constrains the context, ensuring the second-model review focuses only on the relevant code or documentation.

What is prompt bundling for cross-validation across AI models?

Prompt bundling combines a prompt plus relevant files into a single, concise request for a second-model advisory pass. It is used to cross-validate design decisions, perform design checks, or get debugging assistance.

Do I need the Oracle CLI to bundle prompts and files for review?

Yes, the Oracle CLI is required to bundle prompts and files for review. You must provide an explicit file selection, choose an engine mode, and can optionally render or copy the output for easy integration.

What's the best way to compare design decisions across different AI platforms?

The best way to compare design decisions is using engine-agnostic validation, which supports both browser and API modes. By bundling a prompt and files into a one-shot request, you can obtain advisory reviews from multiple models like Claude or Gemini.