Claude for Design Leaders: 32 Claude Skills

What these skills are

A design decision, from who holds it to the write-up that proves it, as installable Claude skills.

This is Claude for the part of design leadership that is not drawing screens: working out who actually decides, writing down why the design is the way it is, getting a room to decide instead of redesign, running critique against a bar somebody wrote down, showing what design changed — and then leading a team and making your own case for the next level.

The skills follow the work in order, in seven steps, and each step hands a named artifact to the next. Framing the problem hands over the stakeholder map, what each person actually meant, the problem statement with what is not being solved, and design's plan on one page. Explaining the decision hands over principles that take a stand, two to four real options with their trade-offs, one DACI approver, a rationale doc and a dated decision log. Persuading the room hands over the pre-wire plan, a deck that asks for the decision on slide one, the feedback log that sorts three opposite notes, and the brief for pushing back. Running critique hands over the written quality bar, the ritual, the session brief and twenty comments turned into actions. Proving impact hands over HEART metrics agreed up front, the write-up built on real data, the UX debt register and the investment case. Leading the team hands over the maturity read, the operating rhythm, the delegation board, the hiring loop, a 30-60-90 for a new designer and SBI feedback prep. Growing yourself hands over your first 90 days as a lead, the staff path, a brag document, the promotion case and a leadership case study.

The red line: Claude drafts the rationale, the deck and the write-up from your real work and real data. It never invents a metric, a user quote or a stakeholder's view, and it never rates a designer or a candidate.

Stakeholder views, user quotes and metrics come only from what you paste. Anything missing stays a bracketed placeholder or a visible gap rather than a plausible number, and impact is written as contribution, not attribution. Each skill uses the real mechanics of its method — the power and interest quadrants, the DACI roles, the decision record statuses, the critique roles and formats, HEART goals to signals to metrics, the delegation levels — rather than a generic gather, analyse, recommend.

Work is scored; people are not. Critique and quality skills judge work against the objectives you stated. The feedback skills structure your own notes into situation, behaviour and impact. The hiring loop is a process and a blank form each interviewer fills alone — Claude never scores, ranks or compares a designer or a candidate. Employment, legal, accessibility law and regulatory points come back as questions ending in “check with a qualified adviser”, never as advice. And every output closes with what a named person decides next, and by when.

Each skill is one named method or one artifact: the inputs it needs, where it breaks, a fixed output template, a finish line and the next skill to run. Where a skill fits a Claude surface it says so — Claude Docs for rationale docs, briefs and write-ups, Claude Slides for the review deck, Claude Design to put real options side by side, a Project to keep a decision log or a brag document running across weeks, and connectors such as Figma or your analytics tool to read frames and data. Connectors read; you send every message yourself.

Mechanically, each skill is one folder with a SKILL.md file. The 32 are grouped into the seven steps, and each one gives full value on its own: run one, run a step, or work the whole thing through in the order it arrives.

Download all 32 skills. One zip: ready-to-install skill zips, readable SKILL.md files, and the sources behind every method. Free, no signup. Download the zip · Browse on GitHub

The 32 skills you get

The set follows the work in seven steps: framing the problem with whoever holds the decision, explaining the decision so it stops being re-argued, persuading the room, running critique against a written bar, proving impact from real data — then leading the team and growing yourself.

1 · Frame the problem

1. Stakeholder Map

Use when: You spend your days aligning and still do not know who actually decides

Output: Power and interest grid, goals, constraints and decision rights, facts apart from guesses, engagement plan, first conversation

2. Stakeholder Interview Guide

Use when: A project starts and you are guessing what the VP means by “make it simpler”

Output: Interview goals, semi-structured guide, probes, note template, summary in each person's own words, gaps to verify

3. Problem Framing Brief

Use when: The team was handed a feature and nobody has said what problem it solves

Output: Problem statement, outcome, evidence on hand and missing, constraints, out of scope, “we believe” hypothesis

4. Design Strategy One-Pager

Use when: You are new in a lead role, or leadership asks for design's plan this year

Output: Diagnosis, guiding approach, three to five coherent actions, what we stop, how we will know, owner

2 · Explain the decision

5. Design Principles

Use when: Every review becomes a taste argument because the team has no shared criteria

Output: Four to seven principles that take a stand, the trade-off each settles, “this, not that” examples, conflict check

6. Design Options Trade-off Table

Use when: Stakeholders saw one design and think it is the only one

Output: Two to four real options, criteria from the brief, consequences in words, what each gives up, evidence or gap

7. DACI Decision Framework

Use when: You are responsible for the design but nobody agreed who signs it off

Output: Decision statement, Driver, one Approver, Contributors, Informed, due date, how the outcome is shared

8. Design Rationale Doc

Use when: You keep re-explaining why the design is the way it is

Output: The decision, options and why not, principles applied, evidence or a visible gap, trade-offs, risks, owner

9. Design Decision Log

Use when: A decision from six months ago is re-opened and nobody remembers why

Output: Dated records with context, decision, status and consequences, revisit triggers, a running file

3 · Persuade the room

10. Pre-Wire Plan

Use when: The big review is next week and you do not want the first reaction in the room

Output: Who to see one to one, what each needs and might object to, objections answered in the deck, what changes on a no

11. Design Review Deck

Use when: You present to leadership and the room argues about colours, not the decision

Output: Decision asked on slide one, options and recommendation, real evidence, risks, feedback in scope, next steps

12. Stakeholder Feedback Log

Use when: Three stakeholders gave three opposite notes and the next round undoes the last

Output: Every comment in one table, duplicates merged, conflicts sent to the approver, accept, decline or park, the reply

13. Design Pushback Brief

Use when: You are asked to ship a dark pattern or a change you think will hurt users

Output: The request, user and business risk with evidence, two alternatives, escalation route, how dissent is recorded

4 · Run critique

14. Design Quality Bar

Use when: “Raise the bar” is said every quarter and nobody has written the bar

Output: Review criteria by stage from usability heuristics and WCAG 2.2, what good enough to ship means

15. Design Critique Ritual

Use when: Crit is a calendar slot where the loudest person redesigns the work

Output: Purpose, who comes, critique split from review, format, roles, rules with consequences, failure modes

16. Critique Session Brief

Use when: You present at crit on Thursday and want feedback, not opinions

Output: Objectives, stage, the one question, out of scope, timebox, prompts that turn reactions into questions

17. Critique Notes to Actions

Use when: You left crit with twenty comments and no idea which ones matter

Output: Notes sorted by objective, preference, out of scope or question, what changes and why, the reply to the room

5 · Prove impact

18. HEART Metrics Plan

Use when: The project starts and nobody agreed what design success looks like

Output: Goals, signals and metrics per HEART dimension, baseline or gap flag, data owner, what PM and design watch

19. Design Impact Write-Up

Use when: Leadership asks what design delivered and you have screens, not outcomes

Output: Problem, what design changed, before and after from real data, contribution not attribution, what we learned

20. UX Debt Register

Use when: UX debt piles up because no single fix is ever “important enough”

Output: Each issue with impact, journey stage, frequency, evidence and effort, value against effort plot, short list

21. Design Investment Case

Use when: You need time or budget for design work that never wins against features

Output: The ask, cost of delay from your own figures, cost of delay divided by duration, what stops, decision owner

6 · Lead the team

22. UX Maturity Assessment

Use when: You step into a lead or head role and need to see how design works here

Output: Six levels across strategy, culture, process and outcomes, the evidence for each, the one move to the next level

23. Design Team Operating Rhythm

Use when: The team's week is all meetings and nobody can say which ones decide anything

Output: Every recurring meeting with purpose and output, keep, change or replace, the week and month on one page

24. Delegation Board

Use when: You are a new lead and keep redesigning other people's work

Output: Decision areas, the delegation level agreed for each from tell to delegate, when you step in, review date

25. Design Hiring Loop

Use when: You open a design role and the loop is “chat with the team and see”

Output: Competencies, stages, portfolio prompts, exercise brief, same questions, a blank form each interviewer fills alone

26. Designer 30-60-90 Day Plan

Use when: A designer starts in two weeks and onboarding is a list of tool logins

Output: Before day one, first week, 30, 60 and 90 day goals, buddy, first critique, first real work, check-in questions

27. SBI Feedback Prep

Use when: You must give hard feedback and it keeps coming out as “it felt off”

Output: Situation, behaviour and impact from your notes, labels turned into observations, the intent question, next step

7 · Grow yourself

28. New Design Lead 90-Day Plan

Use when: You just moved from IC to lead, or into a new head of design role

Output: Weeks 1 to 4 listen, 5 to 8 one visible win, 9 to 12 set the rhythm, what you stop doing as an IC

29. Staff Designer Path Map

Use when: You are senior, do not want to manage, and cannot see what staff means here

Output: IC and manager paths side by side, four staff archetypes adapted to design, the gap to your ladder

30. Brag Document

Use when: Review season comes and you cannot remember what you did in March

Output: Running log of projects and effect, decisions led, mentoring, system work, learning, each entry dated with evidence

31. Promotion Case

Use when: You are going for staff or lead and need to show you already work at that level

Output: Your ladder's criteria, evidence mapped to each, scope, impact and influence stories, gaps, who can speak to it

32. Leadership Case Study

Use when: You stopped designing every screen and your portfolio no longer shows your work

Output: Situation, your role, decisions and trade-offs, how you brought people along, outcomes from real data

Where to start, and how the skills chain

Start with the stakeholder map. Almost everything downstream reads it: the DACI sheet turns it into one approver, the pre-wire plan turns it into the three people you see before the review, and the feedback log sends conflicts back to the person the map named. After that there is no compulsory sequence — bring the thing that is in front of you. Review on Tuesday? Start at the Design Review Deck, which will send you back for the options table if you do not have one. Crit this morning? Start at Critique Notes to Actions. Review season in three weeks? Start at the Brag Document. Every skill lists the inputs it needs, so it tells you when the step before it is the one that is actually missing.

They chain. dlead-stakeholder-map says who holds the decision and dlead-stakeholder-interview-guide says what each of them actually meant, which is what dlead-problem-framing-brief needs to state the problem and what is out of scope; dlead-design-strategy-one-pager puts the year on one page. dlead-design-principles gives the team shared criteria, dlead-design-options-tradeoffs puts two to four real options against them, dlead-daci-decision names the one approver, and dlead-design-rationale-doc and dlead-design-decision-log make the answer survive the next six months. dlead-pre-wire-plan and dlead-design-review-deck get the room to decide, dlead-stakeholder-feedback-log sorts what comes back, and dlead-design-pushback-brief is for the request you should not ship. dlead-design-quality-bar is the written bar dlead-critique-ritual, dlead-critique-session-brief and dlead-critique-notes-to-actions all run against. dlead-heart-metrics-plan agrees success up front so dlead-design-impact-writeup has a baseline, and dlead-ux-debt-register and dlead-design-investment-case make the case for the work that never wins against features. Then dlead-ux-maturity-check, dlead-design-operating-rhythm, dlead-delegation-board, dlead-design-hiring-loop, dlead-designer-30-60-90 and dlead-sbi-feedback-prep run the team, and dlead-new-lead-90-day-plan, dlead-staff-path-map, dlead-brag-document, dlead-promotion-case and dlead-leadership-case-study run your own career.

The sources behind each method are in resources/evidence-and-sources.md, inside the pack.

Setup guide

1. Download the pack. One zip: an install folder with 32 ready-to-upload skill zips, a skills folder with the same 32 skills as readable SKILL.md files, and resources/evidence-and-sources.md, which names the source behind each method.
2. Install your skills. In Claude Code, add the marketplace and install design-leaders-pack, and all 32 load at once. In Claude, turn on code execution in Settings, then Capabilities, then go to Customize, then Skills and upload one zip per skill from the install folder. Prefer working from files? Add the SKILL.md files to your Project knowledge instead; it works, just less cleanly. Keep anything about a named colleague or candidate out of Claude memory.
3. Start with the decision in front of you. New in the role, or starting a project? Run the Stakeholder Map with your kickoff notes and the org chart, so every later skill knows who signs off. Already mid-flight? Bring what is actually happening: the review on Tuesday, the crit notes from this morning, the quarter leadership wants numbers for. Each skill lists the inputs it needs and names the next skill to run, so one answer turns into the chain on its own.

Where to start

The quality bar

Every skill in the pack holds the same standard:

• One method, applied properly: the power and interest quadrants, the DACI roles, the decision record statuses, the critique roles and formats, HEART goals to signals to metrics, the delegation levels — not a generic “gather, analyse, recommend”
• Real voices, real numbers: stakeholder views, user quotes and metrics come only from what you paste. Anything missing stays a bracketed placeholder or a visible gap, and impact is written as contribution, not attribution
• Work is scored, people are not: critique and quality skills judge work against stated objectives, feedback skills structure your own notes, and the hiring loop is a process plus a blank form each interviewer fills alone
• Every output ends with a decision: each template closes with what a named person decides next, and by when
• Adviser points stay with an adviser: employment, legal, accessibility law and regulatory points end with “check with a qualified adviser”
• The red line: Claude drafts the rationale, the deck and the write-up from your real work and real data; it never invents a metric, a user quote or a stakeholder's view, and it never rates a designer or a candidate — you fill the forms and you make the call

Who made this

Polar Bear is a people ops consultancy for human-size teams (20 to 200 people). Built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. This pack is the free, self-serve version of how we work.

This pack is for whoever is accountable for the design decision. If you run a design team, a studio or a product organisation and want it working this way — AI carrying the write-ups and the decks so people do the part only people can do — that's what we build with clients.

Meet Pauline. Want your design team working this way? Bring the step you are stuck on and we will pick the skills for it. Book a 30-minute call · Pauline on LinkedIn

Install in Claude Code

Two lines, and every skill in the pack loads at once.

/plugin marketplace add polar-bear-org/claude-skills
/plugin install design-leaders-pack@polar-bear-skills

Using Claude on the web instead? Download the zip and upload each skill from its install folder under Customize → Skills.