The Complete Claude Guide for Product Managers: 42 Claude Skills for the Whole Product Job

What these skills are

The product job, as installable Claude skills.

This is Claude for the whole product job, stage by stage: the direction set, the customers heard, the backlog decided, the work defined, delivered, launched and then proved — plus the weekly rhythm that keeps it all running.

The skills follow the job in the order the work happens, in eight stages, and each stage hands a named artifact to the next. Set direction hands over the vision, the strategy, the impact map, the market size, the north star and the roadmap. Discover hands over the interview guide, the synthesis, the personas, the journey, the jobs and the opportunity tree. Decide hands over the triage board, the scored backlog and the decision record. Define hands over the one-pager, the PR/FAQ, the PRD and the AI feature spec. Deliver hands over the story map, the stories and the sprint plan. Launch hands over the competitive brief, the go-to-market plan and the release notes. Prove hands over the success metrics, the experiment and the post-launch review — and loops back to Discover. Around all eight sits the rhythm of running the product: the weekly update, the retro and the quarter plan.

The red line: real customers are not optional. Claude drafts the documents, the team talks to users, and a named person makes the call.

Each skill is named as an action and keeps one framework inside: the inputs it needs, where it breaks, a fixed output template, a finish line and the next skill to run. No invented facts — no made-up quotes, customers, market numbers, ROI figures or findings; a missing figure becomes a bracketed placeholder or an open question. Work is scored, people are not: the skills score requests, ideas, options, risks and key results, never team members, and personas are composites rather than profiles of real people.

Where a skill fits a Claude surface — Claude Docs, Claude Slides, Claude Design, Projects, connectors, Claude in Chrome — it says so, and beta features are marked beta.

Mechanically, each skill is one folder with a SKILL.md file. The 42 are grouped into the eight stages, and each one gives full value on its own: run one, run a stage, or work the job through in the order it arrives.

Download all 42 skills. One zip: ready-to-install skill zips, readable SKILL.md files, and the sources behind every method. Free, no signup. The same 42 skills are open on GitHub: polar-bear-org/claude-skills.

The 42 skills you get

The set follows the product job, in eight stages: setting direction, discovering, deciding, defining, delivering, launching, proving — and running the product week to week.

1 · Set direction

1. Write the Product Vision

Use when: The team cannot say in one sentence what it is building toward

Output: Vision statement options, Product Vision Board (target group, needs, product, business goals)

2. Write the Product Strategy

Use when: Everything is high priority and leaders set the roadmap by preference

Output: Diagnosis, guiding policy, coherent actions, “we will not” list

3. Map the Impact

Use when: The roadmap is a feature list and nobody can say whose behaviour each item changes

Output: Goal, actors, impacts, deliverables, the deliverables no impact needs marked to cut

4. Size the Market (TAM, SAM, SOM)

Use when: Leadership asks how big this is and the slide needs a number you can defend

Output: Top-down and bottom-up estimates, sourced assumptions, a range, sanity checks

5. Fill the Lean Canvas

Use when: A new product or bet needs its business logic on one page

Output: One-page business logic, riskiest assumptions marked untested

6. Pick the North Star Metric

Use when: The team ships a lot and cannot say which number it is moving

Output: North star, three to five named inputs with definitions, counter-metrics

7. Build the Product Roadmap

Use when: The roadmap is run from the CEO's inbox and every date becomes a promise

Output: Now-next-later board, confidence notes, “not on the roadmap” list, one-page executive version

2 · Discover

8. Prepare the Customer Interview Guide

Use when: You have 30 minutes with a customer and cannot waste it on opinions

Output: Screener lines, story-based questions, probes, consent note, notes template

9. Synthesise the Interviews

Use when: Five interview transcripts and a PRD due Friday

Output: Affinity clusters, need statements with sources, contradictions, open questions

10. Build the Personas

Use when: The team argues about “the user” and everyone means someone different

Output: Composite personas from interview evidence, each claim linked to its source

11. Map the Customer Journey

Use when: Each team owns a slice and nobody sees where customers drop off

Output: Stages, actions, thoughts and feelings, pain points, opportunities, evidence per cell

12. Map the Jobs to Be Done

Use when: A request arrives with no known job behind it

Output: Job statements, circumstances, forces for and against switching, alternatives

13. Fill the Value Proposition Canvas

Use when: Marketing, sales and product each describe the product differently

Output: Customer profile, value map, fit notes, one positioning line every team uses

14. Build the Opportunity Solution Tree

Use when: Discovery keeps getting squeezed and has to fit between sprints

Output: Outcome, opportunities, target opportunity, three ideas, riskiest assumption per idea

3 · Decide

15. Triage the Feature Requests

Use when: Sales promised it, a big client wants it, the CEO forwarded it, all this week

Output: Request log, the job behind each request, build / later / no sort, reply drafts

16. Score the Backlog with RICE

Use when: 40 requests, room for 5, and the loudest voice keeps winning

Output: Scored backlog, confidence notes, out-of-order exceptions, cut list

17. Score with ICE

Use when: Forty growth ideas and one afternoon to pick what to test

Output: Impact, confidence and ease scores, evidence behind each confidence score, top ten to test

18. Rank with WSJF

Use when: Several teams share one backlog and need one order they can defend

Output: Cost of delay parts divided by job size, ranked list, sensitivity check

19. Cost the Delay

Use when: Everything is urgent and you need a no that holds

Output: Urgency profiles, CD3 order, “yes, if” options, reply to the asker

20. Make the Business Case

Use when: You must argue for a feature in finance language and the ROI numbers are guesses

Output: One-page case in five parts, assumptions table with ranges and owners, what would change the answer

21. Set Up the DACI Decision

Use when: Sign-off never comes, or comes and is reversed next week

Output: Driver, approver, contributors, informed, inputs, decision date

22. Keep the Decision Log

Use when: Settled calls keep coming back as if nobody decided

Output: Dated records (context, decision, status, consequences), reopening rule

4 · Define

23. Write the One-Pager

Use when: An idea needs one page before anyone writes a PRD

Output: Problem, who it is for, why now, the bet, success, what we will not do, the ask

24. Write the PR/FAQ

Use when: A prototype is already built and nobody wrote down why

Output: One-page press release, customer FAQ, internal FAQ

25. Write the Prototype Brief

Use when: Leadership saw the AI demo and thinks it ships next week

Output: The belief it tests, pass line set in advance, what it does not prove, gap-to-production list

26. Write the PRD

Use when: The last spec was read five different ways, or the AI draft runs to 12 pages

Output: Problem and evidence, goals, non-goals, numbered testable requirements, open questions, success measures

27. Spec the AI Feature

Use when: Stakeholders expect it to be right every time, and one prompt fix breaks another case

Output: Task, measurable success criteria, acceptable error rate, eval set plan, human fallback, launch bar

5 · Deliver

28. Build the Story Map

Use when: The backlog is 200 tickets and nobody can see the whole product

Output: User activity backbone, walking skeleton, release slices

29. Write the User Stories

Use when: Engineers say the requirements are thin but not what is missing

Output: Stories, Given / When / Then criteria, INVEST check, refinement questions

30. Plan the Sprint

Use when: The sprint is filled to a date promised upstairs and the team never sized the work

Output: Sprint goal as an outcome, capacity by role, items the team selected, definition of done check

31. Run the Sprint Review

Use when: The review is a demo nobody outside the team attends

Output: Increment against the sprint goal, feedback, backlog changes, short key-updates deck

32. Cut Scope with MoSCoW

Use when: The date is fixed and scope has to give

Output: Must / Should / Could / Won't list, Must budget check, Won't-this-time list

6 · Launch

33. Write the Competitive Analysis

Use when: Sales keeps saying “the competitor has it”

Output: Landscape table with sources, win / loss themes, where you differ, what not to copy

34. Plan the Go-to-Market

Use when: The product is ready and sales, support and marketing are not

Output: Launch tier, audiences, positioning line, readiness by team, enablement brief

35. Run the Pre-Mortem

Use when: The launch date was promised before anyone asked what could go wrong

Output: Failure story, reasons list, top risks with owners and signals, go / adjust / stop

36. Write the Release Notes

Use when: Support hears about changes from customers

Output: Customer notes by change type, internal version for support, who is affected

7 · Prove

37. Set the Success Metrics

Use when: You celebrated a release and still cannot say whether it worked

Output: HEART table (goals, signals, metrics), baseline, target, guardrail

38. Design the Experiment

Use when: Two options, a strong opinion on each side, and enough users to test

Output: Hypothesis, variants, primary metric, guardrails, sample and duration check, decision rule

39. Run the Post-Launch Review

Use when: Two weeks after launch customers still use the old workflow

Output: Intended versus actual, why, keep / change list, keep, fix or remove call

8 · Run the product

40. Write the Weekly Update

Use when: Friday goes on an update people skim, then ask the same questions in three meetings

Output: News-first subject line, bottom line and asks, shipped, at risk, decisions by date, one cut per audience

41. Run the Retro

Use when: The retro raises the same problems every sprint and the actions belong to no one

Output: Last action checked, a format for this cycle, themes, one owned change with a check date

42. Plan the Quarter

Use when: The quarter starts, the goals are a feature list, and every date becomes a promise

Output: Objectives and key results, capacity split, bets, what is out, check-in dates

Where to start, and how the skills chain

There is no compulsory sequence. Bring the stage you are stuck on: if nobody can say what the product is building toward, start at Set direction; if the backlog is the problem, start at Decide; if the release is next week, start at Launch. Every skill lists the inputs it needs, so it tells you when the stage before it is the one that is actually missing.

They chain. pmg-write-product-vision and pmg-write-product-strategy set the direction that pmg-build-product-roadmap sequences and pmg-map-the-impact audits item by item, with pmg-size-the-market and pmg-fill-lean-canvas behind the numbers and pmg-pick-north-star-metric behind the one that matters. pmg-prepare-interview-guide gets the calls, pmg-synthesise-interviews turns them into need statements, and pmg-build-personas, pmg-map-customer-journey, pmg-map-jobs-to-be-done and pmg-fill-value-proposition-canvas read them back; pmg-build-opportunity-tree picks the opportunity to go after. pmg-triage-feature-requests sorts the week's incoming, pmg-score-with-rice, pmg-score-with-ice and pmg-rank-with-wsjf order what survives, pmg-cost-the-delay and pmg-make-business-case hold the line when everything is urgent, and pmg-set-up-daci-decision and pmg-keep-decision-log make the call stick. pmg-write-one-pager, pmg-write-pr-faq, pmg-write-prototype-brief, pmg-write-the-prd and pmg-spec-the-ai-feature define the work; pmg-build-story-map, pmg-write-user-stories, pmg-plan-the-sprint, pmg-run-sprint-review and pmg-cut-scope-with-moscow deliver it; pmg-write-competitive-analysis, pmg-plan-go-to-market, pmg-run-pre-mortem and pmg-write-release-notes launch it. pmg-set-success-metrics, pmg-design-the-experiment and pmg-run-post-launch-review say whether it worked, which sends you back to discovery. Around all of it, pmg-write-weekly-update, pmg-run-the-retro and pmg-plan-the-quarter keep the product running.

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 42 ready-to-upload skill zips, a skills folder with the same 42 skills as readable SKILL.md files, and resources/evidence-and-sources.md, which names the source behind every method and where it stops.
2. Install your skills. In Claude Code, add the marketplace and install complete-claude-guide-for-product-managers-pack, and all 42 load at once. In Claude, turn on code execution in Settings, Capabilities, then go to Customize, 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.
3. Start at the stage you are stuck on. There is no setup phase to clear first. Pick the stage that matches this week — direction, discovery, a contested backlog, a spec, a sprint, a launch, a post-launch read — and run the one skill named for it. Each skill lists the inputs it needs and names the next skill to run, so one run turns into the chain on its own.

Where to start

The quality bar

Every skill in the pack holds the same standard, the one we hold when we run our own product:

• One framework, applied properly: each skill uses the real mechanics of its method — the scales, the grid, the formula, the order of the pass — not a generic gather, analyse, recommend
• It tells you where it breaks: every skill says when not to use it, and which skill in the guide fits instead
• Same shape every time: a fixed output template, and a short done-when check before handover
• No invented facts: no made-up quotes, customers, market numbers, ROI figures or findings — missing data becomes a bracketed placeholder or an open question
• Work is scored, people are not: skills score requests, ideas, options, risks and key results, never team members or individual users, and personas are composites rather than profiles of real people
• A named person decides: the skill drafts the document and sets out the trade-off, the call and the send stay with the person who owns them
• The red line: real customers are not optional — Claude drafts the documents, the team talks to users, and a named person makes 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.

The pack carries one product manager's whole job. When you want your whole team working this way, AI carrying the overhead so people do the part only people can do, across product, delivery and everyday operations, that's what we build with clients.

Meet Pauline. Running the whole product job out of one-off chats? Bring the stage 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 complete-claude-guide-for-product-managers-pack@polar-bear-skills

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