skills.rest
majunbao
majunbaoCOMMUNITY
Explore all skills by this author →
GitHub

oracle

Bundle prompts and selected files into a single context-rich request.

Updated Dec 6, 2016
Search Tags:#cli#oracle#session-management#browser#code-context#prompt-bundling#file-attachments
One-click install
npx skills add https://github.com/majunbao/learn --skill oracle-majunbao
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/majunbao/learn/tree/main/openclaw_tags/openclaw-2026.3.2/skills/oracle
Command: npx skills add https://github.com/majunbao/learn --skill oracle-majunbao

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oracle bundles your prompt + selected files into one “one-shot” request so another model can answer with real repo context (API or browser automation). Treat output as advisory: verify against code + tests.

Core Features & Use Cases

  • One-shot context: Combine prompt content with code and docs to provide rich context for AI responses.
  • Sessioned runs: Maintain long-running analyses across browser or API-powered runs with stored sessions.
  • Flexible file attachment: Attach files with glob patterns, include/exclude rules, and built-in defaults to ignore large or irrelevant files.
  • Engine choices: Switch between browser and API backends as needed for the task.

Quick Start

Provide a task in natural language and attach your project files with --file to start a browser-based context-rich run.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle prompts and files to provide repository context for AI?▼

To bundle prompts and files, you combine your natural language prompt with selected project files using file attachment patterns to create a single context-rich one-shot request for the AI model. This allows the AI to answer with real repository context.

What is the best way to run a long-running code analysis across a codebase?▼

The best way to run long-running code analysis is by using sessioned runs that maintain stored sessions across browser or API-powered runs. This enables continuous investigations across codebases, documentation, and pull requests over time.

Can I attach files using glob patterns and include or exclude rules for AI context?▼

Yes, you can attach files using glob patterns with include and exclude rules. The system also uses built-in defaults to automatically ignore large or irrelevant files when bundling your prompt context.

Does the prompt bundling process support both browser and API backends?▼

Yes, the prompt bundling process supports both browser and API backends. You can switch between browser automation and API engines as needed for your specific code analysis or session-enabled tasks.

What are the limitations of using one-shot context requests for AI code analysis?▼

The main limitation is that the output must be treated as advisory. You should always verify the AI-generated responses against your actual code and tests because the context is bundled statically for the request.

Related Skills

Explore More Software Engineering Skills→
forranyforrany

ai-config-architect

Generate platform-aware AI tooling configurations for Claude Code and Cursor.

Community
Advanced
steipetesteipete
385.7k

oracle

Bundle prompts and files into a single one-shot request for repository-context aware answers.

Community
Advanced
hammadparveezhammadparveez

oracle

Guide oracle CLI usage for bundling prompts and files with AI models.

Community
Intermediate
gensparxgensparx
3

oracle

Bundle prompts with selected files and directories for AI model context.

Community
Advanced