ralph

Convert PRD markdown or text documents into Ralph's prd.json format.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/rhinkle/ralph-explained --skill ralph-rhinkle
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/rhinkle/ralph-explained/tree/main/scripts/ralph/skills/ralph
Command: npx skills add https://github.com/rhinkle/ralph-explained --skill ralph-rhinkle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts PRDs into Ralph's prd.json structure so the Ralph autonomous agent can execute requirements as executable user stories.

Core Features & Use Cases

  • Converts a PRD into a prd.json containing project, branchName, description, and userStories.
  • Enables one-iteration-per-story design by structuring each user story for small, self-contained changes.
  • Produces a ready-to-run prd.json artifact suitable for Ralph's workflow.

Quick Start

Provide a PRD markdown document and ask to convert it to prd.json for Ralph.

Frequently Asked Questions about ralph

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

FAQPage Schema
How do I convert a PRD markdown document into a JSON format for autonomous agents?

The prd.json conversion process requires a PRD markdown or text document as input. The Skill parses your requirements to ensure a valid JSON structure suitable for the autonomous agent workflow.

What is included in the prd.json structure generated for the Ralph autonomous agent?

The prd.json structure includes the project name, branchName, description, and a list of ready-to-run userStories. Each user story is structured for small, self-contained changes to enable one-iteration-per-story execution.

Can I use text documents instead of markdown to generate executable user stories?

Yes, you can use plain text documents instead of markdown to generate executable user stories. The Skill accepts both markdown and text formats to parse requirements into the prd.json structure.

Does the prd.json output include steps for typechecking and verification?

Yes, the prd.json output includes typecheck steps and optional verification steps within the user stories. This ensures the generated executable user stories are ready to run within the Ralph workflow.

What is the best way to structure PRD requirements for one-iteration-per-story execution?

The best way to structure requirements for one-iteration-per-story execution is to convert your PRD into prd.json, which formats each user story as a small, self-contained change ready for the autonomous agent to process.