Skip to content
Back to skills

Biz Prd

ASecurity

Writes a PRD from an idea, meeting notes, or existing code — problem, goals/non-goals, user stories, FR/NFR, success metrics, MoSCoW priority, and acceptance criteria. Use when user says "PRD", "product requirements document", "code to PRD", or "feature spec".

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 9, 2026
ai-agentsgosecurityperformance

Security analysis

A100/100

Scanned October 9, 2026

npx -y skills add meshcode-ai/skills --skill biz-prd --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Biz Prd?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Biz Prd
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/meshcode-ai-biz-prd/badge)](https://www.skillsdirectory.com/skills/meshcode-ai-biz-prd)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: biz-prd
description: Writes a PRD from an idea, meeting notes, or existing code — problem, goals/non-goals, user stories, FR/NFR, success metrics, MoSCoW priority, and acceptance criteria. Use when user says "PRD", "product requirements document", "code to PRD", or "feature spec".
license: MIT
metadata:
  source: "alirezarezvani/claude-skills (MIT) · product-team/code-to-prd"
  category: biz
---

# PRD (Product Requirements Document)

**Fixed structure**: ① Problem/opportunity (whose, why now) ② Goals / non-goals (explicitly state what's out of scope — without them you get scope creep) ③ User stories (_As a ~ I want ~ so that ~_) ④ Functional requirements FR (each = ID + Given/When/Then) ⑤ Non-functional requirements NFR (performance, security, availability — with numbers) ⑥ Success metrics (measurable before/after, ≤3) ⑦ Dependencies, risks, open questions.

**Decision criteria**: requirements must be verifiable — never "fast", say "within p95 200ms". A PRD without non-goals must not be approved. Prioritize with MoSCoW (Must/Should/Could/Won't) — Must contains only survival conditions for this release; the combined Should+Could total stays at or below 2× the Must count. For NFRs, specify only those that deviate from the team's defaults (listing everything is noise).

Code → PRD: read existing code, issues, and conversations and deserialize them into the structure above — promoting implicit decisions in the code (error policies, defaults) into explicit requirements.

Quality gate: does each FR contradict a non-goal, are NFRs testable numbers, can success metrics be measured on a dashboard — if any answer is no, reject and fix. A PRD is a record of decisions, not an essay (decisions left as open items come back as rework during implementation).

## Output
Full PRD (markdown), MoSCoW priority table, open questions & risks list, approval checklist.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…