uniffy-io/uniffy / .agents/skills/create-prd
create-prd
VerifiedCreate a Product Requirements Document from the current conversation and save it under `.agents/local/plans/`. Use when the user asks for a PRD, product specification, or requirements document for an Uniffy feature.
Create PRD: Generate Product Requirements Document
Generate a PRD from the current conversation context. Use the output filename supplied with the request, or PRD.md when none is supplied, and write it under .agents/local/plans/.
Structure
Adapt depth to the available information; every section below, in order:
- Executive Summary - overview (2-3 paragraphs), value proposition, MVP goal
- Mission - mission statement, 3-5 core principles
- Target Users - personas, technical comfort, needs and pain points
- MVP Scope - In Scope (
- [x]) vs Out of Scope (- [ ]), grouped by category - User Stories - 5-8 stories ("As a [user], I want [action], so that [benefit]") with concrete examples
- Core Architecture & Patterns - for Uniffy features: domain-driven vertical slices, ConnectRPC proto-first API, multi-tenant
organization_idscoping, URN + search requirements (architecture.md), and the permission model (access_mode+baseline_role+ContentMember, see.agents/rules/permissions.md) - Tools/Features - detailed feature specifications
- Technology Stack - reference the stack table in
.agents/rules/architecture.md; list new dependencies with versions - Security & Configuration - authn/authz approach, env + per-org settings tiers, both product targets (cloud + self-hosted, see AGENTS.md)
- API Specification - ConnectRPC services, methods, proto messages, example payloads
- Search Integration - if adding a content type, reference the Search Integration Checklist in
architecture.md - Success Criteria - measurable functional requirements and quality indicators
- Implementation Phases - 3-4 phases with goals, deliverables, validation criteria
- Future Considerations - post-MVP enhancements
- Risks & Mitigations - 3-5 key risks with specific mitigations
- Appendix - related documents, key dependencies, structure (if applicable)
Process
- Extract: review the whole conversation; capture explicit requirements, implicit needs, constraints, and success criteria. If critical information is missing, ask before generating.
- Synthesize: organize into the sections, fill reasonable assumptions where details are missing, keep terminology consistent.
- Write: professional and action-oriented; markdown throughout; concrete examples over abstractions; no emojis.
- Confirm: reply with the file path, a brief content summary, assumptions made, and suggested next steps (review, then the
plan-featureandexecuteskills).
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |