ralph-tui-create-json_skill

This skill converts PRD text into a flat prd.json with user stories and quality gates for ralph-tui execution.
  • TypeScript

1.5k

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill subsy/ralph-tui --skill ralph-tui-create-json

  • SKILL.md10.1 KB

Overview

This skill converts product requirement documents (PRDs) into a ralph-tui compatible prd.json file. It extracts quality gates, parses user stories and acceptance criteria, sets dependencies and priorities, and outputs a flat JSON ready for autonomous execution. The JSON adheres to ralph-tui schema rules so the agent can run stories iteratively.

How this skill works

The skill scans the PRD for a Quality Gates section and pulls universal and UI-specific gates. It parses each user story into a single JSON entry (US-001, US-002, ...), appends quality gates to acceptance criteria, assigns priority based on dependency order, and emits a flat root-level object with name, branchName, description, and userStories. It validates against schema anti-patterns (no wrappers, no tasks array, uses passes boolean) and writes the file to ./tasks/prd.json by default.

When to use it

  • You have a PRD in markdown or plain text and want a runnable prd.json for ralph-tui
  • You need consistent acceptance criteria with project-wide quality gates appended
  • You want user stories sized for single-agent iterations
  • You need dependency ordering (schema → backend → UI) encoded for ralph-tui
  • You want a ready-to-run file for ralph-tui run --prd ./tasks/prd.json

Best practices

  • Keep each user story small enough to finish in one agent iteration (2–3 sentence scope)
  • Include a clear Quality Gates section in the PRD or provide commands when asked
  • Derive branchName from the feature in kebab-case and prefix with ralph/
  • Use explicit, verifiable acceptance criteria — avoid vague language
  • Order stories by technical dependency and set dependsOn arrays accordingly

Example use cases

  • Convert a feature PRD into prd.json so ralph-tui can implement database, API, and UI stories sequentially
  • Append CI/lint/typecheck gates to all stories to enforce build quality during automation
  • Split a large feature into right-sized stories and generate IDs/priorities automatically
  • Produce a prd.json for a UI-heavy feature that also includes browser verification gates
  • Create a tasks/prd.json file next to PRD markdown for immediate ralph-tui execution

FAQ

The skill will prompt for required commands or default to a sensible gate like npm run typecheck; include project-specific gates to avoid missing checks.

How are story IDs and priorities assigned?

IDs are sequential (US-001, US-002, ...). Priority is based on dependency order: 1 = highest, with schema first, backend next, UI last.

Where is the prd.json saved?

Default output is ./tasks/prd.json next to the PRD files, but you can specify a different path for ralph-tui to consume.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational