How does it affect the Ralph Wiggum status.json file?

It automatically updates the status field to 'reset' and records the current timestamp, signaling to the Ralph system that the workspace has been cleared.

When is the best time to use Ralph Reset?

You should use it when your AI agent is stuck, when the codebase is in an inconsistent state, or when tests are failing due to leftover artifacts from previous attempts.

Does this skill work with any Git repository?

Yes, it can be used in any Git-managed repository, though it is specifically optimized for workflows using the Ralph Wiggum autonomous development pattern.

Is the ralph-reset operation reversible?

No, this is a destructive operation that performs a hard Git reset and clean. Any uncommitted changes or untracked files will be permanently lost.

Will I be prompted before the reset occurs?

Yes, the skill includes a mandatory confirmation step requiring a 'yes' response before any files are deleted or changes are discarded.

Ralph Wiggum Reset

Name: Ralph Wiggum Reset
Author: stavarengo

bystavarengo

0•

开发者工具

Restores your repository to its last clean state by discarding uncommitted changes and removing untracked artifacts during AI development loops.

Ralph Reset is a specialized safety and maintenance utility designed for developers utilizing the Ralph Wiggum autonomous development technique. When an AI iteration enters a broken state, fails tests due to environmental noise, or becomes stuck in a logic loop, this skill provides a controlled 'panic button' to revert the repository to the last known good commit. By combining a hard Git reset with a deep clean of untracked files and updating the Ralph Wiggum status tracking, it ensures your autonomous coding agent always has a fresh, stable foundation to begin its next task.

主要功能

01Deep cleaning of all untracked files and directories

02Automated status tracking updates in docs/ai/ralph/status.json

03Hard reset to the last committed state (HEAD)

04Interactive safety confirmation to prevent accidental data loss

050 GitHub stars

06Instant recovery from inconsistent or broken AI-generated states

使用场景

01Clearing failed experimental changes to return to a stable baseline

02Ensuring a clean environment before starting a new Ralph Wiggum iteration

03Breaking an AI agent out of an unproductive or recursive iteration loop

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add stavarengo/ralph-wiggum-loop ralph-reset

For use in Claude.ai and ChatGPT

Download Skill