What is the 'hat' system in Ralph Orchestrator?

The hat system uses specialized AI personas—Planner, Builder, Reviewer, and Verifier—to manage different stages of the development process autonomously.

What is the difference between code tasks and runtime tasks?

Code tasks are structured files used to define requirements, while runtime tasks are tracked in .agent/tasks.jsonl to manage the agent's progress during an active run.

Can I use Ralph Orchestrator with existing GitHub issues?

Yes, Ralph can fetch GitHub issues via the CLI, convert them into specifications, and generate an autonomous implementation plan to solve them.

How does Ralph Orchestrator handle code quality?

It implements 'backpressure,' which prevents the agent loop from proceeding unless the implementation passes all specified tests, linting, and type-checking requirements.

Ralph Orchestrator

Name: Ralph Orchestrator
Author: mshuffett

bymshuffett

0•

Productivity & Workflow

Coordinates specialized AI agents in an autonomous coding loop to implement complex features from specifications and PRDs.

Ralph Orchestrator is a robust autonomous coding framework designed to enhance Claude's capabilities through a specialized multi-agent 'hat' system. By utilizing distinct personas like the Planner, Builder, Reviewer, and Verifier, it breaks down high-level requirements or GitHub issues into manageable tasks and executes them within a continuous feedback loop. This skill is particularly effective for PRD-driven development, providing built-in quality gates (backpressure) and an evolutionary meta-orchestrator that refines implementation strategies based on performance analysis.

Key Features

01Multi-agent 'hat' system for specialized planning, building, and verification

02Automated GitHub issue integration for rapid feature implementation

03PRD-driven workflow that maps specifications to structured implementation plans

040 GitHub stars

05Evolutionary meta-orchestrator that identifies and evolves winning coding strategies

06Configurable quality gates and backpressure to enforce testing and linting standards

Use Cases

01Implementing complex software features directly from a Product Requirement Document (PRD)

02Running parallel AI coding approaches to find the most efficient solution for a specific problem

03Resolving multi-part GitHub issues by automatically generating and executing sub-task plans

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mshuffett/dotfiles ralph-orchestrator

For use in Claude.ai and ChatGPT

Download Skill