skills.rest
wangqianCAI
wangqianCAICOMMUNITY
Explore all skills by this author →
GitHub

oracle

Bundle project source code and prompts into a unified context for AI analysis.

Updated Jun 18, 2026
Search Tags:#automation#debugging#context#refactoring#oracle#ai-development
One-click install
npx skills add https://github.com/wangqianCAI/OBI --skill oracle-wangqiancai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/wangqianCAI/OBI/tree/main/skills/oracle
Command: npx skills add https://github.com/wangqianCAI/OBI --skill oracle-wangqiancai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @steipete/oracle.

What problem does it solve?

This Skill solves the challenge of providing AI models with sufficient, relevant repository context for complex debugging, refactoring, or design tasks without hitting token limits or losing focus.

Core Features & Use Cases

  • Context Bundling: Automatically aggregates prompt instructions and selected source files into a single, coherent payload.
  • Browser-Based Long-Thinking: Supports long-running sessions via browser automation for deep, multi-step reasoning tasks.
  • Use Case: When you need to refactor a complex module, use this Skill to bundle the module, its dependencies, and your specific requirements into a single request for a high-end model to generate a comprehensive implementation plan.

Quick Start

Use the oracle skill to bundle all files in the src directory and ask the model to review the current implementation for potential performance bottlenecks.

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 debugging and refactoring tasks?▼

To bundle repository context for AI debugging, this Skill aggregates selected source files and prompt instructions into a unified payload. It automatically manages file globbing and token estimation to prevent context loss during complex code analysis.

What is the best way to analyze a complex codebase without exceeding AI model token limits?▼

The best way to analyze a codebase without exceeding token limits is using context bundling. This approach estimates token counts and aggregates only relevant source files, ensuring high-precision AI analysis stays within model constraints.

Do I need a CLI tool to manage source code context for browser-based AI sessions?▼

Yes, you need the oracle CLI tool to manage source code context for browser-based AI sessions. It handles file globbing, token estimation, and session persistence required for long-running, multi-step reasoning tasks.

Can I use context bundling for long-running architectural design reviews in a browser?▼

Yes, you can use context bundling for architectural design reviews in a browser. It supports long-running sessions via browser automation, allowing high-end AI models to conduct deep, multi-step reasoning over your project source code.

How does project source code bundling work for AI code generation?▼

Project source code bundling works by automatically aggregating your selected files and prompts into a single coherent payload. This unified context allows advanced AI models to generate comprehensive implementation plans and perform complex refactoring.

Related Skills

Explore More Software Engineering Skills→
me2Docme2Doc
6

oracle

Bundle repository files and prompts into a context payload for AI models.

Community
Advanced
zhaozengqing4364-bitzhaozengqing4364-bit

ai-slop-cleaner

Clean up AI-generated code by removing dead code and duplication.

Community
Intermediate
eleva-labseleva-labs

dev-feature

Organize multi-session feature development with structured folders and persistent documentation.

Official
Intermediate
aaronbassettaaronbassett
4

devs:code-review

Automate code reviews with analysis, checklists, and reports for multi-language projects.

Community
Advanced