Product Planner

---
name: product-planner
description: Map user types, goals, flows, screens before coding.
argument-hint: "[description of the app idea, or empty for interactive intake]"
---

The business-analyst phase for greenfield app development. Maps personas, journeys, screens, and features BEFORE any code is written.

This skill fills the gap between "I have an app idea" and "here's the blueprint." It produces a structured journey document that feeds directly into `composure:blueprint`'s greenfield mode as Phase 00.

> **Currently open for beta** — licensing for this feature is being finalized. Full access while in beta.

## Progress Tracking

This skill uses TaskCreate for progress tracking. Before starting work:
1. Create one task per major step using TaskCreate
2. Set each task to `in_progress` when starting it (TaskUpdate)
3. Mark `completed` when done
4. Write deliverables to files, not inline — inline text is for communication only

## Content Loading

Load each step through the fetch command (handles caching, decryption, and auth):

```bash
"$HOME/.composure/bin/composure-fetch.mjs" skill design-forge product-planner {step-filename}
```

**Do NOT read cache files directly** — they are encrypted at rest. Always use the fetch command above.

## Steps

| # | File | Purpose |
|---|------|---------|
| 00 | `00-license-check.md` | Beta access check |
| 01 | `01-intake.md` | Parse description, identify user types + goals |
| 02 | `02-persona-mapping.md` | Generate personas with pain points + device prefs |
| 03 | `03-journey-flows.md` | Map flows per persona (happy path, edge, error) |
| 04 | `04-screen-inventory.md` | Extract + deduplicate screens from flows |
| 05 | `05-feature-extraction.md` | Bridge from UX to code: features, entities, endpoints |
| 06 | `06-write-journey-doc.md` | Write consolidated journey document |

## Output

A structured `user-journey.md` document containing:
1. Personas (name, role, goals, pain points, devices)
2. Journey flows (Mermaid diagrams + prose)
3. Screen inventory (name, purpose, personas, navigation)
4. Feature list (name, screens, data entities, API endpoints, dependencies)

This output is implementation-agnostic — it doesn't prescribe any framework. It feeds into `composure:blueprint` for stack selection and architecture decisions.