The whole PM craft, packed into ~68 skills, and the one that made me stop and look

Last week I was reading through a run of repos that pack product work into skills. Some pick one topic and go deep. This one does the opposite: it is the broadest of the bunch.

It is called pm-skills, by Pawel Huryn, the author of the Product Compass newsletter. He packs almost the entire product management craft into around 68 skills across 9 plugins, from setting strategy, running discovery, and researching the market, to analyzing data, executing, and shipping software that AI wrote.

Usually something this broad ends up shallow. But when I actually opened it, it was not, and one skill in particular made me stop and look for a while, because it covers an angle that only recently became necessary in the era where AI writes code for us.

I will tell it in three parts, starting with what it is, then why it is not just a prompt box, and closing with lessons for anyone building products.

Part 1What pm-skills is

It is a marketplace of around 68 Claude skills for PM, organized into 9 plugins, each one a craft within product work.

• The thinking side product strategy, product discovery, market research, setting direction and understanding the market and customers.
• The doing side data analytics and execution, writing PRDs, prioritizing work, running a pre-mortem, analyzing data.
• The go-to-market side go-to-market, marketing, and growth, planning launches and scaling.
• The new AI shipping side a category built for the PM who now has to own software an AI wrote the code for.

What separates it from a bin of prompts is that it has a validator enforcing that every skill has a complete format: names that match the folder, complete file headers, correct cross-references. It is not a case of anyone dropping in whatever they like. This is a genuinely maintained set, not a pile an AI churned out.

Part 2Why it is not just a prompt box

First, the frameworks are named and sourced. Each skill does not just say "go try some discovery." It builds on frameworks that have real owners, from Teresa Torres, Marty Cagan, Alberto Savoia, and Strategyzer, to SWOT, Porter, Ansoff, JTBD, and RICE, with step-by-step guidance. The author put it plainly in the intro:

"Generic AI gives you text. PM Skills Marketplace gives you structure."

That is where the difference lives. Ask for a SWOT once and you get a scaffold with fields to fill, not a broad paragraph of advice you still have to arrange yourself.

Second, one skill was built for the era of AI-written code. The one that made me stop is called intended-vs-implemented. It audits the gap between what the docs say a system should do and what the code actually does. The skill file puts it sharply:

"A linter checks code in a vacuum. It can tell you whether the code is internally consistent, but not whether it does what you intended, because it has no model of your intent. The highest-value security and correctness bugs live in that gap: a permission written down but never enforced, an endpoint documented as cron-only that anyone can call, a field marked public-only that still leaks private data."

This is the angle a PM now needs. Once AI writes code fast, the problem moves from "is the code written correctly" to "does the code do what we intended," and that is a question a person has to ask, not a tool.

Part 3Lessons for anyone building products

What pm-skills says to everyone building products, not only the people who will actually use it:

• Knowing frameworks is no longer scarce. When SWOT, JTBD, and RICE are a second away, what separates the strong from the average is not how many frameworks you know, but choosing the one that fits the problem in front of you.
• Audit intent, not just that it runs. Borrow the idea behind intended-vs-implemented: a pretty document or a passing demo does not mean the thing does what you intended. A person has to read that gap.
• A maintained set is different from a churned-out one. The format validator and the sourced, named frameworks are what make this set trustworthy. When you meet a repo that packs skills, check whether a person maintains it or an AI just photocopied documentation.

Where to start

Try it on the PM work you already repeat.

1. Take one thing you do regularly, such as writing a PRD or running a competitive analysis.
2. Let the skill in the matching category help with the scaffold, then see which framework it builds on and whether that is the right one to use.
3. If it is something AI wrote the code for, apply the intended-vs-implemented idea and ask whether it actually does what the PRD said.
4. Spend the time you save on the scaffold on the decisions, not on doing more of the same work.

pm-skills is a clear picture of what is happening to PM work. The whole craft is becoming something you can call up almost for free. What is left for us is judgment: choosing right, auditing well, and making the call yourself. This is one in a series where I go through repos that pack a craft into skills, and it will close with a capstone that draws the thread through all of them.

Continue the series

• GTM as a whole playbook in 12 skills a business-side repo that packs the go-to-market method into skills.
• gstack by Garry Tan when a YC president writes his taste straight into software.
• product discovery as a conveyor belt a repo that goes deeper on discovery, built on Teresa Torres's way of thinking.
• AI product management, the capstone the pillar that draws the thread through every skill-packed repo in the series.