oracle

Bundle prompts and repository files into a single oracle run.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/estebanrfp/gos --skill oracle-estebanrfp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/estebanrfp/gos/tree/main/skills/oracle
Command: npx skills add https://github.com/estebanrfp/gos --skill oracle-estebanrfp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oracle helps you package the right prompt and repository files so another model can answer with real project context instead of guessing.

Core Features & Use Cases

  • Prompt bundling: Combine a task statement with a tightly scoped file set for one-shot analysis.
  • Engine selection: Choose browser or API execution depending on the model and workflow you want to use.
  • Session management: Reattach to long-running runs instead of repeating the same request.
  • Use case: Use it for debugging, code review, patch planning, and repo investigation when the answer depends on specific source files and configuration.

Quick Start

Use the oracle skill to bundle the most relevant repository files and ask for a verified, context-aware answer to your coding question.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle repository context for AI code review?

To bundle repository context for AI code review, you select the relevant source files and combine them with a task statement to create a single run for context-aware model assistance.

Can I reattach to a long-running prompt session?

Yes, you can reattach to long-running prompt sessions. Session management allows you to reconnect to existing runs instead of repeating the same request.

How do I estimate token usage before sending files to an AI model?

You can estimate token usage before sending files by performing a dry-run token estimation. This requires disciplined file selection to scope the bundled repository context accurately.

Does prompt bundling support both browser and API execution modes?

Prompt bundling supports both browser and API execution modes. You must explicitly configure the engine selection to choose the workflow and model you want to use for repository analysis.

What is the best way to plan patches using repository analysis?

The best way to plan patches using repository analysis is to bundle a tightly scoped file set with your patch statement for one-shot analysis, ensuring the model has the exact project context needed.

Why does my AI model guess the answer instead of using my codebase?

Your AI model guesses the answer instead of using your codebase because it lacks real project context. You must bundle selected repository files with your prompt to enable verified, context-aware assistance.