# Create Complete Product Requirements Documents

Vague product ideas cause unclear scope, weak acceptance criteria, and delivery risk. This skill turns stakeholder input into a structured, measurable Product Requirements Document.

## Install

```bash
npx skillstore add github/prd
```

## Metadata

- Status: approved
- Slug: github-prd
- Skillstore revision: r2
- Version status: missing
- Tree hash: 5ede08eb3df5df1d54518c242ddedb609041d5a9b53033a2a9b3e649e6edadae
- Author: github
- GitHub username: github
- License: MIT
- Repository: https://github.com/github/awesome-copilot/tree/main/skills/prd/
- Ref: c43861a65bb95efcae259cd161c9d6f4dc7eec6f
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 79
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/github-prd
- Manifest: https://skillstore.pages.dev/api/skills/github-prd/manifest

## Capabilities

- Asks discovery questions about the core problem, success metrics, constraints, and timing.
- Transforms product input into an executive summary with a problem statement, proposed solution, and measurable success criteria.
- Defines user personas, user stories, acceptance criteria, and explicit non-goals.
- Documents architecture, integrations, security, privacy, tools, and evaluation requirements.
- Identifies technical risks, dependencies, hidden complexity, and phased rollout milestones.
- Supports software systems and AI-powered features through a consistent PRD structure.

## Use Cases

- Scope a New Product Feature: Convert stakeholder goals into measurable requirements, user stories, acceptance criteria, non-goals, and rollout phases.
- Align Technical Delivery: Document architecture, integrations, security expectations, dependencies, technical risks, and test requirements before implementation.
- Plan an AI Capability: Define model or tool requirements, evaluation metrics, quality thresholds, privacy controls, and failure handling for an AI feature.

## Prompt Templates

### Draft a Feature PRD

```
Create a PRD for [feature]. First ask at least two questions about the problem, target users, success metrics, and constraints.
```

### Turn an Idea into Requirements

```
Convert [product idea] into a scoped PRD. Include measurable outcomes, user stories, acceptance criteria, dependencies, non-goals, and an MVP rollout.
```

### Specify an AI Feature

```
Write a PRD for [AI feature]. Define tools, data handling, evaluation datasets, quality thresholds, latency targets, costs, safeguards, and failure modes.
```

### Develop a Production Roadmap

```
Create a production PRD for [system]. Analyze architecture, integrations, privacy, security, operational risks, testing, observability, dependencies, and phased release criteria.
```

## Limitations

- The workflow requires at least two clarifying questions before drafting a PRD.
- It does not independently validate feasibility, legal compliance, budgets, or market assumptions.
- It creates requirements and plans but does not implement the product or generate project code.
- The fixed PRD structure may require adaptation for an organization's internal approval process.

## Best Practices

- Provide the business problem, target users, constraints, and expected outcome before drafting.
- Replace subjective words with measurable thresholds for performance, quality, adoption, and reliability.
- Review the first draft with product, engineering, design, security, and operations stakeholders.

## Anti Patterns

- Do not request a final PRD while withholding essential context or refusing discovery questions.
- Do not present unverified assumptions about technology, budgets, compliance, or deadlines as confirmed requirements.
- Do not omit non-goals, acceptance criteria, dependencies, or measurable success metrics.

## Security Audit

- Audited at: 2026-07-23T18:06:32.72\+00:00
- Summary: All six static findings are false positives caused by Markdown fences, inline code formatting, or ordinary prose. SKILL.md contains no executable scripts, command invocation, prompt injection, or intent-level security concern.

## Stats

- Views: 192
- Downloads: 63
- Favorites: 0
- Popularity score: 0