oracle

Bundle prompts with selected files for context-rich model-assisted tasks.

Updated Jan 16, 2026
One-click install
npx skills add https://github.com/czue/clawdbot --skill oracle-czue
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/czue/clawdbot/tree/main/skills/oracle
Command: npx skills add https://github.com/czue/clawdbot --skill oracle-czue

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oracle enables bundling a model prompt with the exact files it should consider during a run, creating context-rich, reproducible AI results and reducing token waste.

Core Features & Use Cases

  • Prompt + File Bundling: Attach relevant project files to a single prompt so models reason with real context.
  • Engine & Session Management: Choose between browser or API engines and manage long-running sessions for complex tasks.
  • Attachment Patterns & Safety: Control which files to include or exclude and apply best practices for secure prompts.

Quick Start

Bundle your prompt with the necessary files and run oracle using the browser engine to generate context-aware results.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I attach files to a prompt for better model context?

You can bundle a model prompt with selected project files using file attachment patterns. This delivers context-rich guidance to the model so it reasons with real codebase data during a run.

Can I choose between browser and API engines for long-running sessions?

Yes, engine management allows choosing between browser or API engines. You can manage long-running sessions to handle complex, file-aware prompting tasks across your codebase.

What is the best way to reduce token waste when running long-context reasoning tasks?

Bundling your prompt with only the necessary files ensures context-rich, reproducible AI results. By controlling file attachments precisely, you reduce token waste from irrelevant codebase context.

Does prompt bundling work with API-driven workflows and structured attachment handling?

Prompt bundling supports both browser-based and API-driven workflows. It handles structured file attachments across codebases to provide exact context for model-assisted tasks.

How do I control which files to include or exclude when attaching context to a prompt?

You control file attachments using specific file patterns. This allows you to include relevant project files and exclude unnecessary ones, applying best practices for secure, context-rich prompting.