agentic-ralph-init

Automate a Ralph loop iterating against acceptance criteria from an end-state spec.

Updated Apr 25, 2026
One-click install
npx skills add https://github.com/kurioscreative/skills --skill agentic-ralph-init
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentic-ralph-init
Source: https://github.com/kurioscreative/skills/tree/main/skills/agentic-ralph-init
Command: npx skills add https://github.com/kurioscreative/skills --skill agentic-ralph-init

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates a Ralph loop that iterates against acceptance criteria from an end-state spec, providing autonomous execution driven by failing checks.

Core Features & Use Cases

  • Acceptance Criteria Loop: Iterates against acceptance criteria from an end-state spec.
  • Autonomous Execution: Agent decides what to implement based on failing checks.
  • Use Case: Ideal for sub-contracts or specs with acceptance criteria, enabling autonomous execution and tracking progress.

Quick Start

Set up a Ralph loop for your spec by running /agentic-ralph-init <path to spec file>.

Frequently Asked Questions about agentic-ralph-init

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

FAQPage Schema
How do I automate spec-driven development against acceptance criteria?▼

You can automate spec-driven development by running a Ralph loop that iterates against acceptance criteria from an end-state spec, allowing an agent to autonomously decide implementation based on failing checks.

What is a Ralph loop for autonomous development execution?▼

A Ralph loop is an autonomous execution cycle that iterates against acceptance criteria from an end-state spec, using failing checks as the signal for the agent to decide what to implement next.

How do I set up an autonomous execution loop for my spec file?▼

You set up an autonomous execution loop by running the init command with a path to your spec file, which starts the Ralph loop to track progress and handle decision-making via Claude.

Do I need bash scripting to run acceptance criteria-driven development loops?▼

Yes, bash scripting is required to handle the Ralph loop iteration, interact with Claude for decision-making, and track progress against your end-state spec acceptance criteria.

Can I use spec-driven development automation for sub-contracts and specific specs?▼

Yes, this autonomous execution approach is ideal for sub-contracts or specs with defined acceptance criteria, enabling the agent to track progress and implement based on the end-state description.

How does the agent decide what to implement during the autonomous execution loop?▼

The agent decides what to implement by evaluating failing checks against the acceptance criteria from the end-state spec, using that signal to drive the next iteration of the Ralph loop.