One-click install
npx skills add https://github.com/Flexasaurusrex/OpenPaw --skill oracle-flexasaurusrex
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/Flexasaurusrex/OpenPaw/tree/main/skills/oracle
Command: npx skills add https://github.com/Flexasaurusrex/OpenPaw --skill oracle-flexasaurusrex

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps users understand and effectively utilize the oracle command-line interface for bundling prompts with files, managing different execution engines, and handling sessions for complex tasks.

Core Features & Use Cases

  • Prompt + File Bundling: Packages prompts and selected files into single requests for LLMs.
  • Engine Selection: Supports API and browser-based execution with configurable options.
  • Session Management: Allows reattachment to long-running or detached tasks.
  • Use Case: When debugging a complex issue in a large codebase, use oracle to bundle your prompt, relevant source files, and error logs, then run it in browser mode to get context-aware suggestions.

Quick Start

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

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle source files and error logs into a single LLM prompt?

You can bundle source files and error logs into a single LLM prompt by using the oracle CLI to attach files with glob patterns, packaging your prompt and selected files together for context-aware suggestions.

What is the best way to manage long-running LLM execution sessions from the command line?

Long-running LLM execution sessions can be managed from the command line using oracle's session management features, which allow you to reattach to detached or ongoing tasks to review results.

How do I preview the token count for a prompt before sending it to an LLM API?

To preview the token count for a prompt before sending it to an LLM API, use the oracle CLI to preview the payload, allowing you to estimate context size for tasks like analyzing all files in a directory.

Can I choose between API and browser-based execution when running prompt engineering tasks?

Yes, you can choose between API and browser-based execution for prompt engineering tasks using oracle's configurable engine selection, letting you run tasks in browser mode or via API based on your needs.

Why does my LLM interaction fail to provide context-aware suggestions for a large codebase?

LLM interactions fail to provide context-aware suggestions for large codebases when prompts lack detailed project context, so you must bundle relevant source files and error logs with your prompt using the oracle CLI.

Does oracle CLI work with glob patterns to exclude test files from file bundling?

Yes, oracle CLI works with glob patterns to exclude test files from file bundling, enabling you to attach only specific files from directories like src while filtering out unnecessary files.