A complete product requirements methodology designed to turn vague ideas into buildable specifications that developers and AI agents can execute reliably.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install afrexai-prd-engine
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install afrexai-prd-engine using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is PRD Engine?
The PRD Engine is more than just a template; it is a full-scale system for engineering product requirements that actually ship. By bridging the gap between high-level stakeholder vision and granular technical execution, this skill ensures that every feature is backed by evidence and a clear definition of success. Using Openclaw Skills like this one helps teams avoid common pitfalls like scope creep and unmeasurable goals. It provides a structured framework for discovery, documentation, and the creation of user stories that are ready for both human developers and AI coding agents.
This methodology focuses on technical truth and actionable documentation. It moves beyond simple text descriptions by incorporating YAML-based data structures for user stories and automated quality rubrics. Whether you are managing a human engineering team or leveraging autonomous agents, the PRD Engine provides the rigorous structure necessary for modern software development lifecycles.
PRD Engine Use Cases
- Turning a vague product idea into a buildable technical specification
- Writing comprehensive PRDs for new features, products, or major architectural refactors
- Reviewing and improving existing requirements before sprint planning to reduce wasted cycles
- Breaking down complex epics into right-sized, INVEST-compliant user stories
- Preparing high-context specifications for AI coding agents such as Claude Code, Cursor, or Copilot
How PRD Engine Works
- Validation: Begin with a Discovery Brief to validate the problem statement, identify affected personas, and establish primary success metrics.
- Documentation: Generate a structured PRD using the standardized template, covering everything from scope exclusions to technical design.
- Story Decomposition: Break the solution into granular user stories using the INVEST methodology to ensure they are small and testable.
- Agent Optimization: Adapt stories for AI execution by adding explicit file paths, context windows, and terminal verification commands.
- Quality Scoring: Use the built-in rubric to score the PRD across dimensions like problem clarity and technical feasibility.
- Lifecycle Management: Track the requirement from draft status through approval, implementation, and post-launch review.
PRD Engine Setup
To implement this methodology in your workflow, start by creating the necessary project context files that Openclaw Skills require for execution. You can initialize your documentation structure using the following commands:
# Create the project context for AI agent awareness
touch PROJECT_CONTEXT.md
# Initialize a new PRD for a specific feature
touch feature-prd.md
Ensure your environment supports YAML and Markdown parsing to get the most out of the structured story formats and discovery briefs provided by this engine.
PRD Engine Data Schema & Taxonomy
The PRD Engine organizes data through a series of specialized schemas designed for both human readability and machine processing.
| Schema Type | Format | Primary Metadata Components |
|---|---|---|
| Discovery Brief | YAML | Problem statement, evidence, impact metrics, constraints |
| PRD Document | Markdown | Solution overview, key decisions, edge cases, release plan |
| User Story | YAML | Narrative, acceptance criteria, priority, size, dependencies |
| AI Agent Story | YAML | File paths to touch, terminal commands, verification logic |
| Project Context | Markdown | Tech stack, directory mapping, coding conventions |
PRD Engine Advanced Features
- Agent-Optimized Formats: Specialized story structures that include explicit file paths and verification commands to prevent AI hallucinations.
- Dependency Pyramid: A logical ordering system that ensures backend logic and schemas are built before UI components.
- Weighted Quality Rubric: A quantitative scoring system (0-100) to determine if a PRD is ready for development or requires more discovery.
- INVEST Validation: Strict criteria for user stories ensuring they are Independent, Negotiable, Valuable, Estimable, Small, and Testable.
- Post-Launch Review Cycle: Integrated templates for measuring actual performance against target metrics 30 days after shipping.
SKILL.md
Loading
Related Openclaw Skills
Incident Postmortem Generator
Transform messy incident notes into structured, blameless postmortem reports using industry-standard SRE best practices.
Post-Mortem & Incident Review Framework
A comprehensive framework for conducting blameless incident reviews to identify root causes and prevent recurring systemic failures.
Portfolio Risk Analyzer
A comprehensive investment risk management system that analyzes positions, calculates complex metrics, and stress tests portfolios using real-time web data.
Policy Writer
A professional-grade document generator for creating structured, compliant internal business policies across various departments.
Presentation Mastery: Complete Slide Design & Delivery System
A comprehensive AI-driven system for building persuasive presentations, covering everything from audience analysis to delivery coaching.
Pricing Optimizer
A sophisticated AI agent that applies value-based frameworks and behavioral psychology to maximize revenue and conversion through optimized pricing tiers.
Featured*
