Free Product Design AI Skills Pack
A helpful little AI skills pack to direct AI tools to walk through all seven layers of product design, so the decisions underneath the screen actually get made.
Earlier this week I was pointed towards Jamie Mill's excellent new Free Product Design Skills AI Pack, a helpful little AI skills pack to direct AI tools to walk through all seven layers of product design, so the decisions underneath the screen actually get made — from reality to problem space to solution space.
Jamie's AI skills cover all parts of good UX work — orientation, observed behavior, the domain, user needs, product and service strategy, conceptual model, interaction structure and flow and surface. For Claude Code, Cursor, Codex and other AI tools.
Product Design AI Skills Pack diagram, strategy, and opportunity trees. Source. Large view.
Better Outcomes Through Guided AI #
Unguided AI delivers generic, non-specific and rarely relevant outcomes. Guided AI is much more aware of the limitations and boundaries it has to respect to be useful.
So it's fantastic to see more tooling trying to guide AI with a healthy dose of human handholding. That means that as designers, we need to thoroughly document how we think about problems and approach them, but also how we iterate on solutions and prioritize them — and what works and fails in real-world scenarios.
Oldie but goodie: a helpful way to approach and solve design challenges, put together by Jamie Mills, building from Jesse James Garrett's legendary work. Large view.
Now, every team works differently, and every team needs to customize skills to make them work for them, in an environment where their design work lives. This doesn't happen by default — it must be intentional.
Good AI Design Work Needs A Good Design System #
To customize a team's design skills well, we need to meticulously document not just what we do, but also things that are often obscured by layers of experience, opinions, politics and internal knowledge.
We also need to invest in design systems, so decisions find their way to be accessible and AI-ready. And with that, we must also understand how decisions are actually made, and communicate to AI our established ways of working where our design must live.
Meet a practical guide on how to reduce drifts, minimize mistakes, maintain context and improve the quality of AI-generated prototypes.
Now that's a new designer's job, and yet again just another sign of how valuable strategically-minded designers will be for years to come. Recent example of the AI prototyping process, put together by Atlassian's Lewis Healey and Kylor Hall (video deep-dive).
Free Useful Guides To Claude For Designers #
Getting started with Claude, Codex or other tools for design or research might seem daunting at first, but it appears way more complicated than it actually is. I've found a few good resources to get started — and perhaps you'll find them useful, too.
Very thorough and very helpful. Free practical guide for designers on Claude Code, by Tommaso Nervegna.
Useful Resources #
- AI Skills Pack: Claude etc., by Jamie Mill
- Atlassian's Deep Dive: Design, Systems & Impact, by Lewis Healey and Kylor Hall
- Claude Code For Designers: A Practical Guide, by Tommaso Nervegna
- Claude Cowork for Designers, by Ryan Rumsey
- Practical Guide to Agentic AI, Design Systems & Figma, by Christine Vallaure de la Paz
- Free Design Skills For Claude, by Marie-Claire Dean
- Inclusive Design Skills for AI Agents, by Mc Dean
- AI-Powered Designer/Developer Collaboration, by Luke Wroblewski
- 100 AI Skills For Product Managers, by Paweł Huryn
