prd

Generate structured Product Requirements Documents with user stories, technical specifications, and risk analysis.

1|Updated Aug 7, 2025
One-click install
npx skills add https://github.com/zzafergok/arktos --skill prd-zzafergok
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prd
Source: https://github.com/zzafergok/arktos/tree/main/.agent/skills/prd
Command: npx skills add https://github.com/zzafergok/arktos --skill prd-zzafergok

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a vague product idea into a concrete, stakeholder-ready specification is slow and error-prone. This Skill guides an AI through a structured discovery interview and produces a complete PRD following a strict schema, so requirements are measurable and unambiguous. ## Core Features & Use Cases - Discovery-First Workflow: Forces clarifying questions about the core problem, success metrics, and constraints before drafting anything. - Strict PRD Schema: Outputs Executive Summary, User Stories with Acceptance Criteria, AI System Requirements, Technical Specifications, and Risks & Roadmap. - Quality Standards: Enforces concrete, measurable requirements (e.g., latency targets, precision benchmarks) instead of vague language. - Use Case: A product manager asks the AI to plan an intelligent search feature; the Skill interviews them, then drafts a PRD with KPIs, user flows, evaluation strategy, and a phased rollout plan. ## Quick Start Ask the AI to write a PRD for your new feature idea and answer its discovery questions about the problem, metrics, and constraints.

Frequently Asked Questions about prd

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a PRD for a new software feature?▼

Start with a discovery phase covering the core problem, success metrics, and constraints, then draft using a strict schema: executive summary, user stories with acceptance criteria, technical specifications, and risks with a phased roadmap.

What should a PRD for an AI-powered feature include?▼

An AI feature PRD should include tool and API requirements, an evaluation strategy with benchmarks and pass rates, plus standard sections like user stories, integration points, and security considerations.

How to make product requirements measurable instead of vague?▼

Replace subjective terms like fast or intuitive with concrete criteria, such as returning results within 200ms on a 10k record dataset or achieving at least 85% Precision@10 in benchmark evaluations.

When should I not write a PRD without more information?▼

Never draft a PRD without asking at least two clarifying questions about the problem, metrics, or constraints. If the tech stack is unspecified, ask the user or explicitly label it as TBD.

What are non-goals in a product requirements document?▼

Non-goals explicitly define what the project will not build, protecting the timeline from scope creep. They belong in the User Experience and Functionality section alongside user stories and acceptance criteria.