FDE PRD Writer
Turn an approved POC into an implementable, testable specification
xukun0821
@xukun0821
Install
$ openclaw skills install @xukun0821/fde-prd-writerFDE PRD Writer | POC Specification Handoff
Turn an aligned customer problem and POC agreement into a PRD package that engineering can implement, QA can accept, and downstream teams can continue delivering.
Confirm upstream readiness
Before starting, confirm the presence of:
- The Customer Problem-Discovery Package from
fde-problem-discovery. - The POC Engagement Charter from
fde-engagement-charter.
When either is missing, state the gap and return to the corresponding upstream stage. Do not fill the PRD with unvalidated assumptions.
Writing process
- Lock the problem and outcome: Extract target users, real workflow, business outcomes, success criteria, and explicit non-goals.
- Design the minimum closed loop: Define scope, functions, exceptions, permissions, data, and dependencies around user tasks. Replace vague terms such as “intelligent” and “easy to use” with observable behavior.
- Write testable specifications: Number each requirement. Define trigger, preconditions, main flow, exception flow, acceptance criteria, and evidence.
- Complete downstream handoffs: Identify the inputs and open risks required by Deployment Architecture, Agent Skill Design, POC Run, and Adoption and Value.
- Run quality checks: Use references/prd-quality.md. Rewrite or mark as open anything that cannot be implemented, tested, or traced.
Method selection principles
Treat KANO, SWOT, JTBD, user stories, and Mermaid as decision and communication tools:
- Use a method only when it improves a scope, priority, risk, or workflow decision.
- Convert every conclusion back into concrete evidence, requirements, and acceptance criteria.
- Use Mermaid to clarify complex flows, states, or system relationships, not as decoration.
See references/method-selection.md for method selection and references/upstream-downstream-contracts.md for lifecycle boundaries.
Load references/user-stories.md, references/test-scenarios.md, andreferences/prd-style.md only when needed. Interviewing, discovery, and product strategy belong to Problem Discovery or Productization and are not the default starting point for this skill.
Output
Use templates/prd-skeleton.md to produce the POC PRD Specification Handoff Package. Default to English in this edition. Produce another language only when the user explicitly requests it.
Before delivery, run node scripts/validate-traceability.js <prd.md>to check structural traceability and orphan identifiers across FR → AC → TS. The script detects identifier and relationship gaps only; it cannot decide whether a requirement, acceptance criterion, or test is correct for the business.
Boundary
This skill does not replace Problem Discovery, the POC Charter, Deployment Architecture, Agent implementation, real execution, or value review. Its responsibility is to give those downstream activities a reliable, traceable, and acceptable specification.
Top skills in this category
self-improving agent
@pskoettCaptures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude rea
Skill Vetter
@spclaudehomeSecurity-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
Self-Improving + Proactive Agent
@ivangdavilaSelf-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
Proactive Agent
@halthelobsterTransform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Now with WAL Protocol, Working Buffer, Autonomous Crons, and battle-tested patterns. Part of the Hal Stack 🦞
Agent Browser
@matrixyHeadless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection