← all repositories
deanpeters/product-manager-prompts

AI Prompts for PMs That Ask Questions, Not Fill Blanks

These prompts exist because treating ChatGPT like a magic template machine is why your user stories still sound like Mad Libs.

997 stars Python LLMOps · EvalLearning
product-manager-prompts
Not currently ranked — collecting fresh signals.
star history

What it does

This repo holds over 50 markdown prompts for ChatGPT, Claude, Gemini, and similar assistants, covering bread-and-butter PM work like user stories, competitive analysis, and Jobs-to-be-Done research. Instead of handing you fill-in-the-blank templates, each prompt engineers a conversation where the AI asks clarifying questions and you make the strategic calls. Files are organized into directories such as /prompts/ for execution when context is already clear, /prompt-generators/ for guided discovery, and even /storytelling/ for stakeholder presentations.

The interesting bit

The secret sauce is not the prompt text but the HTML comment blocks wrapped around it—metadata that explains the methodology, warns about missing context, and teaches you how to work with AI. Because GitHub’s rendered preview hides these comments, the README demands you view the raw source to see the “instructor’s notes,” forcing you to learn why the prompt works while you use it. It is a surprisingly thoughtful pedagogical trick for a collection of copy-paste tools.

Key highlights

  • Built on established frameworks like Jobs-to-be-Done and Value Proposition Canvas rather than generic output.
  • Explicitly splits /prompts/ (for when you have context and need output) from /prompt-generators/ (for when you need the AI to facilitate discovery).
  • Includes /skeletons/ for reverse-engineering how the prompts actually function under the hood.
  • The submission guide requires contributors to test prompts in multiple assistants and solve real PM problems they have personally faced.
  • MIT-licensed community build; the author explicitly frames this as human-led decision making with AI as a thinking aid.

Caveats

  • The repository assumes you already know product management fundamentals; these are collaboration accelerants, not coursework.
  • Directories like /vibes/ and /resumes-resignations-reactions/ are experimental or humorous, so utility varies.
  • The “view raw code” requirement is easy to miss on first visit, meaning newcomers may literally not see the most instructive parts of the files.

Verdict

Bookmark this if you are a working PM who wants to draft faster and pressure-test strategy through structured AI dialogue. Look elsewhere if you want a fully automated PM replacement or a beginner textbook.

Frequently asked

What is deanpeters/product-manager-prompts?
These prompts exist because treating ChatGPT like a magic template machine is why your user stories still sound like Mad Libs.
Is product-manager-prompts open source?
Yes — deanpeters/product-manager-prompts is an open-source project tracked on heatdrop.
What language is product-manager-prompts written in?
deanpeters/product-manager-prompts is primarily written in Python.
How popular is product-manager-prompts?
deanpeters/product-manager-prompts has 997 stars on GitHub.
Where can I find product-manager-prompts?
deanpeters/product-manager-prompts is on GitHub at https://github.com/deanpeters/product-manager-prompts.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.