deanpeters/Product-Manager-Skills
Product Management skills framework built on battle-tested methods for Claude Code, Cowork, Codex, and AI agents.
What it solves
This project provides a structured library of 77 professional product management (PM) frameworks and skills to prevent AI agents from producing generic, low-quality output. It ensures that both the human PM and the AI agent share a professional foundation of reasoning, failure modes, and judgment, reducing the need for repetitive prompting and guesswork.
How it works
The library consists of 77 SKILL.md files organized into three tiers of complexity:
- Workflow Skills: End-to-end processes that can take days or weeks (e.g., running a full discovery cycle).
- Interactive Skills: Guided discovery tools that use an "Adaptive Decision Ladder" to ask 3-5 targeted questions and then provide tailored recommendations based on the context.
- Component Skills: Templates for specific deliverables (e.g., writing a user story with acceptance criteria).
These skills can be integrated into various AI platforms such as Claude (Desktop/Web/Code), OpenAI Codex, ChatGPT, Cursor, and n8n, or run locally via a Streamlit playground.
Who it’s for
Product Managers who want to upgrade their AI agents' capabilities or learn the professional reasoning behind common PM frameworks.
Highlights
- Comprehensive PM Coverage: Includes skills for framing, strategy, stakeholder alignment, customer discovery, prioritization, roadmapping, and product lifecycle/EOL (End-of-Life) management.
- Adaptive Decision Ladder: Interactive skills guide the user through a decision process rather than just dumping a framework.
- Platform Agnostic: Works with any agent that can read structured markdown, with specific installation guides for major AI tools.
- Multi-tier Architecture: Organizes work from high-level workflows down to specific component templates.
Related
- Project
- Project
- Project
- Project
- Project