oracle

Bundle prompts with selected files and directories for LLM context.

2|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/FatBy/DD-OS --skill oracle-fatby
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/FatBy/DD-OS/tree/main/skills/oracle
Command: npx skills add https://github.com/FatBy/DD-OS --skill oracle-fatby

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the process of providing relevant code context to LLMs for tasks like code generation, debugging, and analysis, by intelligently bundling prompts with selected files.

Core Features & Use Cases

  • Context Bundling: Packages prompts with specified files and directories for LLM processing.
  • Engine Flexibility: Supports both API and browser-based execution engines.
  • Use Case: When debugging a complex issue in a large codebase, use this Skill to bundle your error message and relevant source files into a single request for an LLM to analyze and suggest a fix.

Quick Start

Use the oracle skill to preview the token count for a task involving all files in the src directory.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle code files to provide context for an LLM prompt?

You can bundle code files into LLM context by using a CLI to attach specified files and directories to your prompt. This packages your code and instructions together for code generation, debugging, or analysis tasks.

What is the best way to provide LLM context for debugging a large codebase?

Providing LLM context for debugging a large codebase is best done by bundling error messages and relevant source files into a single request. This approach ensures the LLM analyzes the specific code alongside the issue.

Can I preview token usage before sending my code context to an LLM?

Yes, you can preview token usage before sending code context to an LLM. The CLI supports token count previewing, allowing you to estimate the size of your bundled prompt and attached files before execution.

Does this CLI support both API and browser automation engines for LLM tasks?

Yes, this CLI supports both API and browser automation engines for LLM tasks. This engine flexibility allows you to choose between direct API calls and browser-based execution depending on your workflow needs.

How do I attach files using globs when bundling prompts for code analysis?

You can attach files using globs when bundling prompts for code analysis by specifying glob patterns in the CLI. This allows you to selectively include multiple files and directories that match your desired code context.

Do I need to install a specific binary to manage LLM sessions and code context?

Yes, you need to install and configure the 'oracle' binary to manage LLM sessions and code context. This binary is required for engine execution and enables features like session management and context bundling.