PM Skills Marketplace: The AI-Powered Toolkit That Gives Product Managers Real Structure
You’ve got a new product idea. You open your AI assistant and ask for help — and what you get back is a wall of generic text. No framework. No structure. No rigor. Just words.
That’s the gap PM Skills Marketplace was built to fill. It’s an open-source collection of 68 product management skills, 42 chained workflows, and 9 installable plugins that turn any compatible AI assistant into a structured product decision-making machine. From initial discovery through strategy, execution, launch, and growth — every stage of the product lifecycle is covered.
This guide breaks down what it is, how it works, what’s inside, and how to get started.
What Is PM Skills Marketplace?
At its core, PM Skills Marketplace is an AI skill marketplace designed specifically for product managers. It encodes proven PM frameworks from experts like Teresa Torres, Marty Cagan, Alberto Savoia, and Dan Olsen into individual “skills” — modular knowledge units that your AI assistant can draw on during a conversation.
The key difference from using a generic AI? Structure.
Generic AI gives you text. PM Skills Marketplace gives you a framework. Each skill walks you through a well-established methodology step by step — whether that’s assumption mapping during discovery, a Business Model Canvas during strategy, or a PRD template during execution.
The project was originally built for Claude Code and Claude Cowork, but the skill files use a universal format that works with other tools too, including Gemini CLI, OpenCode, Cursor, and Kiro.
How It Works: Skills, Commands, and Plugins
The system is organized into three layers. Understanding how they relate to each other is the key to using it effectively.
Skills — The Building Blocks
A skill is the most basic unit. Each one gives your AI assistant a specific domain knowledge module, analytical framework, or guided workflow. For example, the prioritization-frameworks skill teaches the AI to reference ICE, RICE, MoSCoW, Kano, and five other frameworks whenever you discuss backlog prioritization.
Skills load automatically when relevant to your conversation — you don’t need to invoke them manually. If you want to force-load a specific skill, you can call it directly using the format /plugin-name:skill-name.
Commands — Chained Workflows
A command is a user-triggered workflow activated with a /slash call. Commands chain one or more skills together into an end-to-end process. Here’s what that looks like in practice:
-
/discoverruns: brainstorm ideas → identify assumptions → prioritize assumptions → design experiments -
/write-prdguides you from a feature concept to a structured Product Requirements Document
Commands are designed to flow into each other. When one completes, the system suggests the most relevant next command — you just follow the prompts.
Plugins — Installable Packages
A plugin groups related skills and commands into a single installable package. The marketplace includes 9 plugins, each covering a distinct PM domain. Installing the marketplace gives you all 9 at once.
Here’s how the three layers relate:
| Layer | What It Does | Triggered How |
|---|---|---|
| Skill | Single framework or knowledge module | Loads automatically; can be forced with /skill-name |
| Command | Multi-skill workflow | Manually with /command-name |
| Plugin | Bundle of skills and commands | Installed once; everything inside activates as needed |
How to Install PM Skills Marketplace
Installation depends on which AI platform you’re using. Here are the four supported paths.
Option 1: Claude Cowork (Best for Non-Developers)
This is the simplest path — all graphical, no terminal required:
-
Open Claude Cowork and click Customize (bottom-left corner) -
Navigate to Browse plugins → Personal → click + -
Select Add marketplace from GitHub -
Enter: phuryn/pm-skills
All 9 plugins install automatically.

Option 2: Claude Code (CLI)
For developers comfortable with the command line:
# Step 1: Add the marketplace
claude plugin marketplace add phuryn/pm-skills
# Step 2: Install each plugin
claude plugin install pm-toolkit@pm-skills
claude plugin install pm-product-strategy@pm-skills
claude plugin install pm-product-discovery@pm-skills
claude plugin install pm-market-research@pm-skills
claude plugin install pm-data-analytics@pm-skills
claude plugin install pm-marketing-growth@pm-skills
claude plugin install pm-go-to-market@pm-skills
claude plugin install pm-execution@pm-skills
claude plugin install pm-ai-shipping@pm-skills
Option 3: Codex CLI (OpenAI)
Codex reads the same marketplace file natively — no conversion needed:
# Step 1: Add the marketplace
codex plugin marketplace add phuryn/pm-skills
# Step 2: Install the plugins you want
codex plugin add pm-toolkit@pm-skills
codex plugin add pm-product-strategy@pm-skills
codex plugin add pm-product-discovery@pm-skills
codex plugin add pm-market-research@pm-skills
codex plugin add pm-data-analytics@pm-skills
codex plugin add pm-marketing-growth@pm-skills
codex plugin add pm-go-to-market@pm-skills
codex plugin add pm-execution@pm-skills
codex plugin add pm-ai-shipping@pm-skills
One important difference: Codex’s plugin system doesn’t support /slash commands directly. Instead, describe the workflow steps in plain language. For example:
“Run product discovery on my idea: brainstorm options, map assumptions, prioritize the risky ones, then design experiments — pause between each step.”
You can also ask Codex to convert frequently used command files into native Codex skills for faster access.
Option 4: Other AI Assistants (Skills Only)
If you use Gemini CLI, OpenCode, Cursor, or Kiro, you can copy the skill folders into the appropriate directory. Commands (/slash style) are Claude-specific and won’t work on other platforms.
| Tool | How to Use | What Works |
|---|---|---|
| Gemini CLI | Copy skill folders to .gemini/skills/ |
Skills only |
| OpenCode | Copy skill folders to .opencode/skills/ |
Skills only |
| Cursor | Copy skill folders to .cursor/skills/ |
Skills only |
| Kiro | Copy skill folders to .kiro/skills/ |
Skills only |
Batch copy example (OpenCode):
for plugin in pm-*/; do
mkdir -p .opencode/skills/
cp -r "$plugin/skills/"* .opencode/skills/ 2>/dev/null
done
All 9 Plugins Explained
This is the heart of the marketplace. Each plugin covers a specific product management domain, and together they span the full PM lifecycle.
1. Product Discovery (pm-product-discovery)
Purpose: Take an idea from spark to validated experiment.
This is one of the largest plugins — 13 skills and 5 commands — covering everything from early-stage brainstorming through experiment design.
Key skills:
| Skill | What It Does |
|---|---|
brainstorm-ideas-existing |
Multi-perspective ideation (PM, Designer, Engineer) for existing products |
brainstorm-ideas-new |
Ideation for new products in initial discovery |
brainstorm-experiments-new |
Design lean startup pretotypes (Alberto Savoia methodology) |
identify-assumptions-existing |
Surface risky assumptions across Value, Usability, Viability, and Feasibility |
identify-assumptions-new |
Identify risks across 8 categories including Go-to-Market, Strategy, and Team |
prioritize-assumptions |
Rank assumptions with an Impact × Risk matrix and suggest experiments |
opportunity-solution-tree |
Build Teresa Torres’s OST — outcome → opportunities → solutions → experiments |
interview-script |
Generate structured interview guides with JTBD probing questions |
summarize-interview |
Turn transcripts into JTBD insights, satisfaction signals, and action items |
Commands:
-
/discover— Full discovery cycle -
/brainstorm— Multi-perspective ideation (ideas|experiments × existing|new) -
/triage-requests— Batch-analyze and prioritize feature requests -
/interview— Prepare an interview script or summarize a transcript -
/setup-metrics— Design a product metrics dashboard
Example usage: You have an idea for an “AI meeting summarizer for remote teams.” Run /discover AI-powered meeting summarizer for remote teams and the system walks you through the entire flow from divergent thinking to experiment design.
2. Product Strategy (pm-product-strategy)
Purpose: Vision, business models, pricing, and competitive analysis.
12 skills and 5 commands covering the complete strategic toolkit.
Highlights:
-
product-strategy — A 9-section Product Strategy Canvas (vision through defensibility) -
startup-canvas — A combined product strategy + business model canvas, designed as an alternative to the Lean Canvas and BMC for new products -
value-proposition — 6-part JTBD value proposition template (Who, Why, What before, How, What after, Alternatives) -
pricing-strategy — Pricing models, competitive analysis, willingness-to-pay, and price elasticity -
swot-analysis, pestle-analysis, porters-five-forces, ansoff-matrix — Classic strategic analysis frameworks, each available as a standalone skill
Commands:
-
/strategy— Build a complete 9-section Product Strategy Canvas -
/business-model— Explore models (lean|full|startup|value-prop|all) -
/value-proposition— Design a value proposition using the JTBD template -
/market-scan— Combine SWOT + PESTLE + Porter’s Five Forces + Ansoff Matrix for macro analysis -
/pricing— Design a pricing strategy with competitive analysis
3. Execution (pm-execution)
Purpose: The day-to-day PM toolkit.
This is the largest plugin with 16 skills and 11 commands, covering PRDs, OKRs, roadmaps, sprint planning, retrospectives, release notes, stakeholder management, and more.
Key skills:
-
create-prd — 8-section PRD template -
brainstorm-okrs — Team-level OKRs aligned with company objectives -
outcome-roadmap — Transform feature lists into outcome-focused roadmaps -
sprint-plan — Sprint planning with capacity estimation and risk identification -
stakeholder-map — Power × Interest grid with tailored communication plans -
prioritization-frameworks — Reference guide to 9 frameworks (Opportunity Score, ICE, RICE, MoSCoW, Kano, and more) -
strategy-red-team — Adversarial stress-testing of any plan: surface load-bearing assumptions and rank by cheapest test -
pre-mortem — Risk analysis using a Tigers/Paper Tigers/Elephants classification
Commands include: /write-prd, /plan-okrs, /transform-roadmap, /sprint (plan|retro|release), /pre-mortem, /red-team-prd, /meeting-notes, /stakeholder-map, /write-stories (user|job|wwa), /test-scenarios, /generate-data.
4. Market Research (pm-market-research)
Purpose: User research and competitive intelligence.
7 skills and 3 commands.
Skills: User personas, market segmentation, user segmentation from feedback data, customer journey maps, market sizing (TAM/SAM/SOM), competitor analysis, and sentiment analysis.
Commands:
-
/research-users— Build personas, segment users, map customer journeys -
/competitive-analysis— Analyze the competitive landscape -
/analyze-feedback— Sentiment analysis and segment insights from user feedback
Example: Got 200 NPS responses from last quarter? Run /analyze-feedback and attach the file. The system extracts sentiment patterns and thematic insights automatically.
5. Data Analytics (pm-data-analytics)
Purpose: Data analysis tools for PMs.
3 skills and 3 commands — small in number but high in utility:
-
sql-queries — Generate SQL from natural language (BigQuery, PostgreSQL, MySQL) -
cohort-analysis — Retention curves, feature adoption, and engagement trends by cohort -
ab-test-analysis — Statistical significance, sample size validation, and ship/extend/stop recommendations
Commands: /write-query, /analyze-cohorts, /analyze-test
6. Go-to-Market (pm-go-to-market)
Purpose: From identifying your first market segment to executing a launch.
6 skills and 3 commands.
Skills: Full GTM strategy, beachhead segment identification, ideal customer profiles (ICPs), growth loop design, GTM motion evaluation (product-led, sales-led, etc.), and competitive battlecards.
Commands:
-
/plan-launch— Full GTM strategy from beachhead to launch plan -
/growth-strategy— Design growth loops and evaluate GTM motions -
/battlecard— Create a sales-ready competitive battlecard
7. Marketing & Growth (pm-marketing-growth)
Purpose: Product marketing and growth metrics.
5 skills and 2 commands.
Skills: Marketing ideas, product positioning, value proposition statements, product naming, and North Star metrics.
Commands:
-
/market-product— Brainstorm marketing ideas, positioning, value props, and names -
/north-star— Define your North Star Metric and supporting input metrics
8. Toolkit (pm-toolkit)
Purpose: PM utilities beyond core product work.
4 skills and 5 commands.
Skills: PM resume review (against 10 best practices), NDA drafting, privacy policy generation (GDPR/CCPA compliant), and grammar/logic checking.
Commands: /review-resume, /tailor-resume, /draft-nda, /privacy-policy, /proofread
9. AI Shipping (pm-ai-shipping)
Purpose: Audit and document AI-generated code.
This plugin solves a problem that’s becoming increasingly common: AI agents write code fast but leave no record of intent — what the system should do, who can do what, where secrets live, which rules are actually verified. This kit restores reviewability.
2 skills and 5 commands.
Skills:
-
shipping-artifacts — Defines the durable documentation set every AI-built app needs: architecture, user/permission flows, variables/secrets inventory, and test-coverage maps -
intended-vs-implemented — A method for finding the gap between what docs say a system should do and what the code actually does
Commands:
-
/ship-check— Turn a vibe-coded repo into a reviewer-ready shipping packet -
/document-app— Reverse-engineer a codebase into system documentation -
/derive-tests— Map documented intent to test coverage and identify gaps -
/security-audit-static— Static security audit with trust boundary mapping and evidence-backed risk reporting -
/performance-audit-static— Find over-fetching, missing indexes, and caching opportunities
Quick Start: Which Command Should You Try First?
Not sure where to begin? Match your situation to the right entry point:
| What You Need | Command |
|---|---|
| Validate a new product idea | /discover |
| Get strategic clarity | /strategy |
| Write a PRD | /write-prd |
| Plan a product launch | /plan-launch |
| Define your North Star metric | /north-star |
These commands chain naturally. After /discover completes, the system suggests /strategy or /write-prd as the logical next step. Just follow the prompts.
The Methodology Behind the Skills
These aren’t arbitrary frameworks. Each skill is grounded in the published work of recognized product management experts:
| Expert | Key Work | Skills Inspired |
|---|---|---|
| Teresa Torres | Continuous Discovery Habits | Opportunity Solution Trees, continuous discovery workflows |
| Marty Cagan | INSPIRED and TRANSFORMED | Product management philosophy embedded across plugins |
| Alberto Savoia | The Right It | Pretotyping experiments for new products |
| Dan Olsen | The Lean Product Playbook | Lean product methodology |
| Roger L. Martin | Playing to Win | Strategic choice frameworks |
| Ash Maurya | Running Lean | Lean Canvas, lean execution |
| Strategyzer | Business Model Generation and Value Proposition Design | BMC and VP Design skills |
| Christina Wodtke | Radical Focus | OKR practices |
| Anthony Ulwick | Jobs to Be Done | JTBD-based skills throughout |
| Alistair Croll & Benjamin Yoskovitz | Lean Analytics | Analytics frameworks |
| Sean Ellis | Hacking Growth | Growth hacking methodologies |
| Maja Voje | Go-To-Market Strategist | GTM strategy skills |
The project was curated by Pawel Huryn from The Product Compass Newsletter.
Pairing with PM Brain
PM Skills Marketplace has a companion project called PM Brain — a “second brain” for product managers. It’s a folder of plain Markdown files on your laptop. Claude reads them before answering, writes to them after, and organizes them weekly. No vector database, no cloud sync, no agent memory tricks.
When used together, PM Brain provides your personal knowledge base while PM Skills Marketplace provides the analytical frameworks. Your AI assistant gets both at once.
Known Issues and Things to Know
Windows Compatibility
Users on Windows may experience instability with Claude Cowork’s VM startup. A PowerShell scheduled task can auto-restart the CoworkVMService when it stops, resolving roughly 90% of the issues. For the remaining cases, manually start the “Claude” service via services.msc.
Licensing
The project is released under the MIT License — free to use, modify, and distribute.
Platform Compatibility Summary
| Platform | Skills | Commands |
|---|---|---|
| Claude Cowork | ✅ | ✅ |
| Claude Code | ✅ | ✅ |
| Codex CLI | ✅ | ❌ (use plain language instead) |
| Gemini CLI | ✅ | ❌ |
| OpenCode | ✅ | ❌ |
| Cursor | ✅ | ❌ |
| Kiro | ✅ | ❌ |
Frequently Asked Questions
How is this different from just asking ChatGPT or Claude product questions directly?
The difference is structure. A general-purpose AI gives you a text response. PM Skills Marketplace makes the AI follow a specific, proven analytical framework step by step — the same frameworks used by experienced PMs at top product companies. You get rigor, not just words.
I’m not a product manager. Is this useful for me?
If your work involves product decisions — as a founder, engineering lead, designer, or anyone shaping what gets built — many of the discovery, strategy, and analysis tools will be relevant. If your role has nothing to do with product work, the applicability is limited.
Does it cost anything?
No. The project is open source under the MIT License.
Do commands work on all platforms?
No. /slash commands are specific to the Claude ecosystem (Claude Code and Cowork). On Codex, Gemini CLI, OpenCode, Cursor, and Kiro, only the underlying skills are available. On Codex, you can describe the workflow steps in natural language to get similar results.
Do I need to manually enable each skill after installing?
No. Skills load automatically when the conversation topic is relevant. To force a specific skill, use /plugin-name:skill-name.
Do I have to install all 9 plugins?
In Claude Cowork, the marketplace installs everything at once. In Claude Code and Codex, you can install individual plugins. That said, workflows often span multiple plugins, so installing the full set is recommended.
The Bottom Line
PM Skills Marketplace takes the product management methodologies that live in books and courses — frameworks from Teresa Torres, Marty Cagan, Alberto Savoia, and others — and encodes them into actionable, AI-invocable skills. It doesn’t replace your judgment. It ensures that when you’re making product decisions, you have a structure to follow.
If you’re already using Claude Code, Claude Cowork, or another compatible AI assistant for product work, this is an open-source resource worth exploring. Start with /discover and validate your next idea with real rigor.

