Ralph Tui PrdSAFE

An agent skill published by subsy.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: ralph-tui-prd
description: "Generate a Product Requirements Document (PRD) for ralph-tui task orchestration. Creates PRDs with user stories that can be converted to beads issues or prd.json for automated execution. Triggers on: create a prd, write prd for, plan this feature, requirements for, spec out."
---

# Ralph TUI PRD Generator

Create detailed Product Requirements Documents optimized for AI agent execution via ralph-tui.

---

## The Job

1. Receive a feature description from the user
2. Ask 3-5 essential clarifying questions (with lettered options) - one set at a time
3. **Always ask about quality gates** (what commands must pass)
4. After each answer, ask follow-up questions if needed (adaptive exploration)
5. Generate a structured PRD when you have enough context
6. Output the PRD wrapped in `[PRD]...[/PRD]` markers for TUI parsing

**Important:** Do NOT start implementing. Just create the PRD.

---

## Step 1: Clarifying Questions (Iterative)

Ask questions one set at a time. Each answer should inform your next questions. Focus on:

- **Problem/Goal:** What problem does this solve?
- **Core Functionality:** What are the key actions?
- **Scope/Boundaries:** What should it NOT do?
- **Success Criteria:** How do we know it's done?
- **Integration:** How does it fit with existing features?
- **Quality Gates:** What commands must pass for each story? (REQUIRED)

### Format Questions Like This:

```
1. What is the primary goal of this feature?
   A. Improve user onboarding experience
   B. Increase user retention
   C. Reduce support burden
   D. Other: [please specify]

2. Who is the target user?
   A. New users only
   B. Existing users only
   C. All users
   D. Admin users only
```

This lets users respond with "1A, 2C" for quick iteration.

If you are Claude Code in an interactive session, you may present these questions with the AskUserQuestion tool instead - same questions and options, still one set at a time. When this skill is driven by `ralph-tui create-prd`, the agent runs non-interactively and nothing can answer a tool call, so ask in plain text with lettered options as above.

### Quality Gates Question (REQUIRED)

Always ask about quality gates - these are project-specific:

```
What quality commands must pass for each user story?
   A. pnpm typecheck && pnpm lint
   B. npm run typecheck && npm run lint
   C. bun run typecheck && bun run lint
   D. Other: [specify your commands]

For UI stories, should we include browser verification?
   A. Yes, use dev-browser skill to verify visually
   B. No, automated tests are sufficient
```

### Adaptive Questioning

After each response, decide whether to:
- Ask follow-up questions (if answers reveal complexity)
- Ask about a new aspect (if current area is clear)
- Generate the PRD (if you have enough context)

Typically 2-4 rounds of questions are needed.

---

## Step 2: PRD Structure

Generate the PRD with these sections:

### 1. Introduction/Overview
Brief description of the feature and the problem it solves.

### 2. Goals
Specific, measurable objectives (bullet list).

### 3. Quality Gates
**CRITICAL:** List the commands that must pass for every user story.

```markdown
## Quality Gates

These commands must pass for every user story:
- `pnpm typecheck` - Type checking
- `pnpm lint` - Linting

For UI stories, also include:
- Verify in browser using dev-browser skill
```

This section is extracted by conversion tools (ralph-tui-create-json, ralph-tui-create-beads) and appended to each story's acceptance criteria.

### 4. User Stories
Each story needs:
- **Title:** Short descriptive name
- **Description:** "As a [user], I want [feature] so that [benefit]"
- **Acceptance Criteria:** Verifiable checklist of what "done" means

Each story should be small enough to implement in one focused AI agent session.

**Format:**
```markdown
### US-001: [Title]
**Description:** As a [user], I want [feature] so that [benefit].

**Acceptance Criteria:**
- [ ] Specific verifiable criterion
- [ ] Another criterion
```

**Note:** Do NOT include quality gate commands in individual story criteria - they are defined once in the Quality Gates section and applied automatically during conversion.

**Important:**
- Acceptance criteria must be verifiable, not vague
- "Works correctly" is bad
- "Button shows confirmation dialog before deleting" is good
- Each story should be independently completable

### 5. Functional Requirements
Numbered list of specific functionalities:
- "FR-1: The system must allow users to..."
- "FR-2: When a user clicks X, the system must..."

Be explicit and unambiguous.

### 6. Non-Goals (Out of Scope)
What this feature will NOT include. Critical for managing scope.

### 7. Technical Considerations (Optional)
- Known constraints or dependencies
- Integration points with existing systems
- Performance requirements

### 8. Success Metrics
How will success be measured?

### 9. Open Questions
Remaining questions or areas needing clarification.

---

## Writing for AI Agents

The PRD will be executed by AI coding agents via ralph-tui. Therefore:

- Be explicit and unambiguous
- User stories should be small (completable in one session)
- Acceptance criteria must be machine-verifiable where possible
- Include specific file paths if you know them
- Reference existing code patterns in the project

---

## Output Format

**CRITICAL:** Wrap the final PRD in markers for TUI parsing:

```
[PRD]
# PRD: Feature Name

## Overview
...

## Quality Gates
...

## User Stories
...
[/PRD]
```

**File naming:** The TUI will save to `./tasks/prd-[feature-name].md`

---

## Example Conversation Flow

**User:** Create a PRD for adding dark mode to the TUI

**Assistant:** I'll help you create a PRD for dark mode. Let me ask a few questions:

1. What is the primary goal?
   A. Reduce eye strain for long sessions
   B. Match system theme preference
   C. Provide visual customization
   D. Other

2. Should the setting persist?
   A. Yes, save to con

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.

[![ModelsAtlas audit: Ralph Tui Prd](https://modelsatlas.com/skills/subsy/ralph-tui-prd/badge.svg)](https://modelsatlas.com/skills/subsy/ralph-tui-prd/)

Not your listing, or something out of date? Claim or update your listing.