ralph-tui-prd_skill
- 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-prd- SKILL.md9.3 KB
Overview
This skill generates a Product Requirements Document (PRD) for ralph-tui task orchestration from a short feature prompt. It interactively asks iterative, lettered clarifying questions (including mandatory quality gates) and produces a machine-friendly PRD wrapped in [PRD]...[/PRD] markers for TUI parsing. The PRD is optimized for conversion to beads issues or prd.json for automated execution.
How this skill works
The skill accepts a feature request and begins an iterative Q&A: 3–5 essential multiple-choice questions per round, always including quality gate commands. After each response it adapts with follow-ups until sufficient context is gathered. Once ready, it emits a structured PRD (overview, goals, quality gates, user stories, functional requirements, non-goals, technical considerations, success metrics, open questions) wrapped in [PRD] markers for ralph-tui tooling.
When to use it
- You need a clear, machine-actionable PRD for a new ralph-tui feature.
- You want user stories and acceptance criteria that can be converted to beads issues or prd.json.
- You must ensure specific CI/lint/typecheck commands are enforced for every story.
- You want iterative clarification to avoid ambiguous requirements before implementation.
Best practices
- Provide a concise feature summary to start (1–2 sentences) to speed iteration.
- Answer lettered options like '1A, 2C' for fast, deterministic clarification rounds.
- Always specify the quality gate commands you expect for CI verification.
- Keep scope small so each user story can be completed by a single AI agent session.
- Flag any known integration points or file paths early to make stories executable.
Example use cases
- Spec out a new TUI command for parallel task orchestration with clear acceptance criteria.
- Create PRD for adding an export-to-json task that integrates with existing task schemas.
- Plan a UI verification flow that requires dev-browser checks plus lint/typecheck gates.
- Define incremental automation stories so each can be converted to a beads task.
FAQ
No—quality gate commands are collected once in the Quality Gates section and applied to every story by the TUI conversion tools.
How many clarification rounds are typical?
Typically 2–4 rounds. The skill adaptively asks follow-ups until it has precise, verifiable acceptance criteria.
