prd-generator

Generate prd.json and progress.txt artifacts from decomposed user stories.

2|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/thesyntaxninja/syntaxninja-dojo --skill prd-generator-thesyntaxninja
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prd-generator
Source: https://github.com/thesyntaxninja/syntaxninja-dojo/tree/main/skills/prd-generator
Command: npx skills add https://github.com/thesyntaxninja/syntaxninja-dojo --skill prd-generator-thesyntaxninja

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the tedious and error-prone task of authoring the structured artifacts required to run the ralph training loop, ensuring prd.json and progress.txt are consistent, verifiable, and placed where the loop expects them.

Core Features & Use Cases

  • Structured PRD generation: Produce prd.json with title, goal, created timestamp, sequential story IDs, story metadata, file lists, verification commands, and a passes flag.
  • Progress tracking: Emit a human-readable progress.txt summary that mirrors prd.json and helps teams and the loop monitor completion.
  • Validation rules: Ensure JSON validity, runnable verification command syntax, and that story dependencies form a DAG to avoid cycles.
  • Use case: After decomposing stories and approving a training loop, run this Skill to scaffold .claude/plugin/ralph/ with prd.json, progress.txt, and notes so the ralph-loop can begin autonomous iterations.

Quick Start

Generate a prd.json and progress.txt from decomposed stories and place them in .claude/plugin/ralph/ so the ralph loop can run.

Frequently Asked Questions about prd-generator

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

FAQPage Schema
How do I generate a prd.json file for an autonomous training loop?

Generate a prd.json file for a training loop by running this Skill after decomposing user stories and approving loop activation; it outputs structured JSON with sequential story IDs, runnable verification commands, and DAG-safe dependencies.

What is the purpose of progress.txt in ralph training cycles?

The purpose of progress.txt in ralph training cycles is to provide a human-readable summary mirroring prd.json, enabling teams and the loop to monitor autonomous iteration completion within the .claude/plugin/ralph/ directory.

How do I scaffold the .claude/plugin/ralph/ directory for loop execution?

Scaffold the .claude/plugin/ralph/ directory by running this Skill on approved decomposed stories; it automatically places the generated prd.json, progress.txt, and notes artifacts exactly where the ralph-loop expects them.

Does the prd-generator validate story dependencies before running a training loop?

Yes, the prd-generator validates story dependencies by ensuring they form a Directed Acyclic Graph (DAG) to avoid cycles, checking JSON validity, and verifying that verification command syntax is runnable.

When should I not use automated PRD generation for progress tracking?

You should not use automated PRD generation before your user stories are fully decomposed and you have explicitly approved loop activation, as the Skill requires these completed steps to produce valid artifacts.