From prd-builder
This skill should be used when conducting PRD interviews, creating product requirements documents, planning new features, documenting bug fixes, or when using commands like "/prd-builder:prd", "/prd-builder:feature", "/prd-builder:bugfix", or "/prd-builder:refine". Provides comprehensive interview frameworks and question templates for building thorough PRDs.
How this skill is triggered — by the user, by Claude, or both
Slash command
/prd-builder:prd-interviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Guide comprehensive interviews to transform rough ideas into actionable Product Requirements Documents — a brief **Launch Framing** phase (thesis + hero flow) followed by structured discovery across 8 categories.
Guide comprehensive interviews to transform rough ideas into actionable Product Requirements Documents — a brief Launch Framing phase (thesis + hero flow) followed by structured discovery across 8 categories.
Apply these to every PRD (skip thesis/hero-flow only for bug fixes). They shape how the interview and document are organized, regardless of stack:
When the user has not specified a stack or architecture, apply the default SaaS profile in
references/default-stack.md as a pre-selected recommended default. That file is the
single source of truth for the profile — stack versions, multi-tenancy, RBAC, auth, MFA,
dependency policy, and payments-out-of-scope. Don't restate the version list elsewhere; point
to it.
/feature: a feature lives inside an existing product, so default to that
product's current stack. Fall back to the full profile only when the host stack is unknown
or the product is greenfield.references/design-review.md); save only the final PRD/architect:design on the saved PRD before tasks — the architecture (docs/architecture/ + docs/adr/) is what taskmanager decomposes and verifies against; size the ceremony to the work and skip it for trivial/low-risk changesAfter the Launch Framing phase, conduct interviews across these 8 categories (adapt based on PRD type):
| Category | Focus Areas | When to Use |
|---|---|---|
| Problem & Context | Pain points, current state, why now | Always |
| Users & Customers | Personas, segments, user journeys | Always |
| Solution & Features | Feature list, MVP scope, priorities | Always |
| Technical Implementation | Architecture, stack (default profile), integrations, dependency vetting | Always |
| Business & Value | ROI, pricing, revenue model | Products, major features |
| UX & Design | Flows, wireframes, accessibility | UI-facing work |
| Risks & Concerns | Dependencies, assumptions, blockers | Always |
| Testing & Quality | Test strategies, acceptance criteria | Always |
When using AskUserQuestion:
Skip categories based on context:
Save interview progress to: .taskmanager/prd-state.json
{
"sessionId": "uuid",
"prdType": "product|feature|bugfix",
"slug": "feature-name",
"startedAt": "ISO timestamp",
"lastUpdatedAt": "ISO timestamp",
"currentCategory": "category-name",
"completedCategories": ["category1", "category2"],
"answers": {
"category-name": {
"question-key": "answer-value"
}
},
"initialPrompt": "User's original description"
}
When /prd-builder:prd or similar is invoked:
.taskmanager/prd-state.jsoncurrentCategorySave PRDs to: docs/prd/prd-{slug}.md
templates/prd-template.md is the authoritative section structure — generate PRDs by
filling it; don't maintain a second copy of the section list here. The ordering principle: the
Product Thesis and Hero Flow lead and the technology follows in service of the hero
flow; scope lives in Features & Requirements (a lean Launch Scope v1 plus a sequenced
Deferred Roadmap); every resolved choice is recorded in Decisions with an owner; Open
Questions holds only what genuinely can't be decided yet.
Include these diagrams where appropriate:
Architecture Diagram:
graph TB
subgraph Frontend
UI[User Interface]
end
subgraph Backend
API[API Layer]
DB[(Database)]
end
UI --> API --> DB
User Flow Diagram:
flowchart LR
A[Start] --> B{Decision}
B -->|Yes| C[Action]
B -->|No| D[Alternative]
The PRD is the what; the architecture is the how. For non-trivial work (a new system, a
new data model/ownership decision, a new service boundary or public contract — anything
expensive to change later), recommend running /architect:design docs/prd/prd-{slug}.md
before tasks. The architect writes docs/architecture/ + docs/adr/ (per scribe's docs
contract), and taskmanager then decomposes and verifies design-conformance against them —
not just against the PRD. Size the ceremony to the work: skip the architecture pass for
trivial/low-risk changes that need no seam or decision. The architect plugin is part of this
suite; if it is not installed, proceed to tasks and note the design step was skipped.
After PRD completion:
/taskmanager:plan with the PRD file pathFeature: User Authentication
├── Setup authentication infrastructure
├── Implement login endpoint
├── Implement registration endpoint
├── Add password reset flow
├── Create authentication middleware
└── Write authentication tests
After generating tasks, ask: "Tasks created. Start autonomous execution?"
If yes, invoke /taskmanager:run to begin implementation.
/prd-builder:prd)Complete interview covering all 8 categories in depth:
/prd-builder:feature)Lighter interview focused on implementation:
/prd-builder:bugfix)Problem-focused documentation:
/prd-builder:refine)Enhance existing PRDs:
Detailed question banks, default profile, and review lenses:
references/question-bank.md - Complete question library organized by categoryreferences/default-stack.md - Default SaaS stack & conventions profile, applied as a pre-selected default when the user has not specified a stackreferences/design-review.md - Dual-lens design review (Experience & Simplicity + First-Principles Engineering) run silently after the draft to redesign the product; only the final PRD is savedPRD output template:
templates/prd-template.md - Full PRD document structurenpx claudepluginhub mwguerra/plugins --plugin prd-builderFetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Applies a firm's KYC/AML rules grid to parsed onboarding records: assigns risk rating, checks required documents, outputs rule outcomes with citations, and routes for escalation.
Generates daily or weekly digests of activity from connected sources (chat, email, docs, tasks, CRM), highlighting action items, decisions, mentions, and project updates.