# Productowner Skill

> Core Product Owner orchestration. Defines PRD acceptance criteria, manages the SDLC lifecycle, and delegates tasks safely to other skills (OpenHands, MetaGPT via local wrappers) without raw execution.

- Skill: `vignesh-s7/productowner-skill` (Agent Skill)
- Install (CLI): `npx skillmds@latest add vignesh-s7/productowner-skill`
- Raw SKILL.md: https://api.skillmd.com/api/skills/vignesh-s7/productowner-skill/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: vignesh-s7 (https://skillmd.com/u/vignesh-s7)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/vignesh-s7/productowner-skill

---


# Product Owner Orchestration (Safe Mode)

## 1. Y-Score Readiness Check
Before proceeding with any feature, you must perform a readiness check against the Y-Score framework to ensure the feature has a clear definition of ready (DoR).

## 2. Gherkin Acceptance Criteria (Given/When/Then)
Always enforce strict BDD/Gherkin acceptance criteria in every PRD.
* **Given** [context]
* **When** [action]
* **Then** [outcome]

## 3. Delegation & Orchestration
Do **NOT** execute bash scripts to orchestrate other agents. Instead, use declarative instructions and safe integrations:
* **Architecture:** Delegate to `system-architect-skill` (Mermaid.js).
* **Engineering:** Delegate to `builder-skill` (Sandboxed OpenHands wrapper).
* **QA:** Delegate to `qa-tester-skill` (Playwright definitions).
* **Security:** Must pass the `cybersec-skill` gate before finalizing.

## 4. Strict Data Minimization (PII/PHI)
**CRITICAL SECURITY RULE:** Never extract, transmit, or retain PII (Personally Identifiable Information) or PHI (Protected Health Information). Always use synthetic, mock data in PRDs and test cases.