ralph-wiggum

Automate spec-driven software development with isolated iterations and on-disk state persistence.

Updated Feb 5, 2026
One-click install
npx skills add https://github.com/joaopiccioni44/Claude-skills --skill ralph-wiggum-joaopiccioni44
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph-wiggum
Source: https://github.com/joaopiccioni44/Claude-skills/tree/main/user-skills/ralph-wiggum
Command: npx skills add https://github.com/joaopiccioni44/Claude-skills --skill ralph-wiggum-joaopiccioni44

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ralph Wiggum automates the process of autonomous, spec-driven software development, enabling iterative task execution with fresh context to ensure clean progress across multiple specs.

Core Features & Use Cases

  • Iterative spec-driven development: the agent works through specs one by one, testing and committing upon meeting all acceptance criteria.
  • Persistent, auditable state on disk: specs, plans, and history are stored for reproducibility and resumption.
  • Guarded autonomy: completion signals are emitted only when criteria are fully satisfied and changes are pushed.

Quick Start

Instruct your AI to run the Ralph Wiggum loop against your set of specs to begin automated development.

Frequently Asked Questions about ralph-wiggum

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

FAQPage Schema
How do I automate spec-driven software development with AI?

Spec-driven development is automated by applying iterative loops with fresh context per iteration, ensuring verifiable completion across multiple specs. The agent tests and commits changes only when all acceptance criteria are fully satisfied.

What is the best way to manage state persistence for autonomous AI coding loops?

State persistence is managed by storing specs, plans, and history directly on disk for reproducibility and resumption. This approach ensures that autonomous AI coding loops maintain auditable progress across multiple development iterations.

How do I set up a spec-driven loop for multi-spec software projects?

Set up a spec-driven loop by defining a set of specs with explicit acceptance criteria for your multi-spec software project, then instruct your AI to run the autonomous loop. It processes tasks incrementally and signals DONE only after all criteria pass.

Can I use autonomous AI coding for incremental task completion across multiple specs?

Yes, autonomous AI coding supports incremental, verifiable completion across tasks for multi-spec software projects. It applies isolated per-iteration context and on-disk state persistence to ensure clean progress through each spec.

Why does the autonomous AI agent only signal DONE after all acceptance criteria pass?

The autonomous AI agent signals DONE only after acceptance criteria pass to enforce guarded autonomy and prevent incomplete task execution. This explicit signaling ensures changes are pushed and verified before moving to the next iteration.

Does the spec-driven development loop require fresh context for each iteration?

Yes, the spec-driven development loop requires fresh context per iteration to ensure clean progress across multiple specs. This isolation prevents context bleed and maintains reproducible, auditable state on disk for incremental task execution.