MCP Servers

PM-Skills MCP

MCP server exposing 40 best-practice product management skills plus workflows and lifecycle tools, instantly accessible to any AI via Model Context Protocol.

README

<h1 align="center"> <a href="https://github.com/product-on-purpose/pm-skills-mcp">PM-Skills MCP</a> </h1>

<h4 align="center">Open-source MCP server exposing 40 best-practice product management skills plus workflows and lifecycle tools, instantly accessible to any AI via Model Context Protocol.</h4>

<a href="https://github.com/product-on-purpose/pm-skills-mcp/issues/new?labels=bug">Report a Bug</a> · <a href="https://github.com/product-on-purpose/pm-skills-mcp/issues/new?labels=enhancement">Request a Feature</a> · <a href="https://github.com/product-on-purpose/pm-skills-mcp/discussions">Ask a Question</a>

Want to customize skills or use slash commands? This server is powered by pm-skills, the open-source skill library you can fork and modify. Browse the full catalog, interactive showcase, and setup guides at product-on-purpose.github.io/pm-skills.

[!IMPORTANT] Maintenance Mode (effective 2026-05-04)

Active development on pm-skills-mcp is paused pending demonstrated demand. The latest release is v2.9.3 (security patch shipped 2026-05-05; embedded catalog frozen at the v2.9.2 build) which remains fully functional and exposes a catalog of 40 PM skills, 11 workflows, and 8 utility tools (59 tools total) via the Model Context Protocol. Security patches and critical bug fixes will continue to be addressed; new skill parity with the upstream pm-skills library is on hold, so the catalog will continue to drift from pm-skills as new skills are added there.

For new users: the file-based install path documented in the main pm-skills repository is under active maintenance and is the recommended path going forward.

To register interest in resumed development: open a GitHub Discussion.

<a href="#the-big-idea">About</a> • <a href="#getting-started">Getting Started</a> • <a href="#usage">Usage</a> • <a href="#project-status">Status</a> • <a href="#contributing">Contributing</a> • <a href="#community">Community</a>

<details> <summary>Table of Contents</summary>

The Big Idea
Getting Started
Usage
Guides
Project Status
Development
Security
Contributing
- How to Contribute
- Reporting Bugs
About
Community

</details>

Quick Start (NPM Global Install)

npm install -g pm-skills-mcp

The Big Idea

PM-Skills MCP is an MCP server that transforms PM-Skills, a collection of 40 best-practice product management skills, into programmatically accessible tools, resources, and prompts for any AI assistant that speaks the Model Context Protocol.

PM-Skills MCP is built on pm-skills, an open-source collection of 40 PM skills organized by the Triple Diamond framework. While pm-skills offers file-based access with slash commands and AGENTS.md discovery, pm-skills-mcp wraps those same skills in an MCP server for programmatic access. The pm-skills library now includes lifecycle tools (Create, Validate, Iterate) and a documentation site with an interactive showcase.

Not sure which to use? See the Comparison section below, or the MCP Setup Guide on the docs site.

One connection. 40 skills. Any MCP client.

Why MCP?

The Model Context Protocol is an open standard that lets AI assistants connect to external tools and data sources. Instead of manually copying skill files or uploading ZIP archives, PM-Skills MCP gives your AI direct, programmatic access to the entire PM skill library.

The Transformation

File-Based Approach	MCP-Powered Approach
Clone repo, navigate to skills	`npx pm-skills-mcp` - done
Copy/paste skill content into chat	AI invokes tools directly
Manual template injection	Automatic template retrieval
Platform-specific slash commands	Universal across all MCP clients
Static markdown files	Dynamic, parameterized tools
Works in one IDE/client	Works everywhere MCP is supported

Key Features

59 MCP Tools - 40 PM skills + 11 workflows + 8 utility tools
MCP Resources - Skill instructions, templates, and examples via URI, with optional persona resources when enabled
3 MCP Prompts - Guided conversation starters for common workflows
11 Workflows - Pre-built multi-skill workflows for common scenarios
81 Automated Tests - Comprehensive test coverage with Vitest
Zero Configuration - Works out of the box with embedded skills
Universal Compatibility - Claude Desktop, Cursor, Continue, and any MCP client
Customizable - Override with your own skill modifications
Lightweight - Minimal dependencies, fast startup
Security Scanning - CodeQL analysis on every push

Built with...

Model Context Protocol - Open standard for AI tool connectivity
PM-Skills - The 40 PM skills this server exposes (docs site)
TypeScript - Type-safe implementation
Vitest - Fast, modern testing framework

Works for...

PM-Skills MCP works with any client that supports the Model Context Protocol. Here's a quick overview:

Platform Compatibility

Platform	Status	Integration Type	Notes
Claude Code	✅ Native	MCP Server	Full tool access via CLI
Claude Desktop	✅ Native	MCP Server	Recommended for best experience
Claude.ai	✅ Native	MCP Server (Projects)	Via MCP integration in Projects
Cursor	✅ Native	MCP Server	AI-powered IDE
VS Code	✅ Native	Via Cline/Continue	Multiple MCP-compatible extensions
Continue	✅ Native	MCP Server	Open-source coding assistant
Cline	✅ Native	MCP Server	VS Code extension
Windsurf	✅ Native	MCP Server	AI-native IDE
GitHub Copilot	🔶 Indirect	Via MCP-enabled tools	Use with Copilot Chat + MCP
OpenCode	🔶 Indirect	Via MCP-compatible config	Use standard stdio MCP server pattern
ChatGPT / Codex	🔶 Manual	Copy skill content	No native MCP support
Any MCP Client	✅ Universal	Protocol-level	Full compatibility

See the Integration Guide for detailed setup instructions for each platform.

Comparison: `pm-skills-mcp` (this repo) vs. `pm-skills`

PM-Skills is available in two complementary forms:

	pm-skills-mcp (this repo)	pm-skills
What it is	MCP server wrapping the skill library	Skill library as markdown files
Access method	`npx pm-skills-mcp`	Git clone, ZIP upload
Setup time	30 seconds	2-5 minutes
Skill invocation	MCP tool calls	Slash commands (Claude Code)
Auto-discovery	MCP protocol (Claude Desktop, Cursor)	AGENTS.md (Copilot, Cursor, Windsurf)
Template access	URI-based resources	Navigate file system
Workflows	Tool-based execution	Manual orchestration
Customization	Set `PM_SKILLS_PATH` to custom folder	Edit files directly
Updates	`npm update pm-skills-mcp`	`git pull`

Use pm-skills-mcp (this repo) when:

You want instant setup with npx pm-skills-mcp
You're using Claude Desktop, Cursor, or any MCP client
You want programmatic tool access without managing files
You prefer consistent interfaces across different AI clients

Use pm-skills (file-based) when:

You prefer slash commands in Claude Code (/prd, /hypothesis)
You want to browse, read, and customize skill files directly
You're using GitHub Copilot, OpenCode, or Windsurf without reliable MCP support
You want to fork and heavily customize skills for your team

Using Both Together: Fork pm-skills to customize skills, then point pm-skills-mcp to your fork:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"],
      "env": {
        "PM_SKILLS_PATH": "/path/to/my/forked/pm-skills/skills"
      }
    }
  }
}

(<a href="#readme-top">back to top</a>)

Getting Started

This section covers quick installation. For a comprehensive walkthrough including core concepts, your first tool invocation, workflows, and troubleshooting, see the Getting Started Guide.

NPM Global Install

npm install -g pm-skills-mcp

Run directly:

pm-skills-mcp

From Source

git clone https://github.com/product-on-purpose/pm-skills-mcp.git
cd pm-skills-mcp
npm install
npm run build
npm start

Quick Start by Platform

<details> <summary>Claude Desktop</summary>

The recommended client for pm-skills-mcp. Add to your Claude Desktop configuration:

Config file location:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json

Configuration:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"]
    }
  }
}

Restart Claude Desktop. You now have access to all 40 PM skills as tools (catalog frozen at the v2.9.2 build; see Maintenance Mode notice above).

</details>

<details> <summary>Claude Code (CLI)</summary>

Add to .claude/settings.json in your project or ~/.claude/settings.json globally:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"]
    }
  }
}

</details>

<details> <summary>Cursor</summary>

Navigate to Settings → Features → MCP Servers, then add:

{
  "pm-skills": {
    "command": "npx",
    "args": ["pm-skills-mcp"]
  }
}

Restart Cursor. The 40 PM skill tools will be available in AI chat (catalog frozen at the v2.9.2 build; see Maintenance Mode notice above).

</details>

<details> <summary>GitHub Copilot (MCP-enabled)</summary>

Configure your Copilot MCP extension/integration with the same stdio server pattern:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"]
    }
  }
}

If your Copilot surface does not currently expose MCP server configuration, use file-based pm-skills with AGENTS.md as the fallback path.

</details>

<details> <summary>OpenCode (MCP-enabled)</summary>

Use OpenCode's MCP server configuration with the standard npx pm-skills-mcp pattern:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"]
    }
  }
}

If MCP is unavailable in your current OpenCode setup, use file-based pm-skills and AGENTS.md.

</details>

<details> <summary>VS Code (Cline / Continue)</summary>

With Cline:

Install Cline from VS Code Marketplace

Open Cline settings and configure MCP server:

{
  "pm-skills": {
    "command": "npx",
    "args": ["pm-skills-mcp"]
  }
}

With Continue:

Install Continue from VS Code Marketplace
Configure MCP in Continue settings

</details>

<details> <summary>Other MCP Clients</summary>

Any MCP-compatible client can use pm-skills-mcp. The general pattern:

Configure an MCP server with command npx and args ["pm-skills-mcp"]
Restart the client
All 40 PM skills become available as tools (catalog frozen at the v2.9.2 build; see Maintenance Mode notice above)

See the Integration Guide for detailed instructions for all platforms.

</details>

Configuration

Environment Variables

Variable	Default	Description
`PM_SKILLS_PATH`	(embedded)	Path to custom skills directory
`PM_PERSONAS_PATH`	(auto)	Path to persona library resources
`PM_SKILLS_FORMAT`	`full`	Default output format
`PM_SKILLS_INCLUDE_EXAMPLES`	`false`	Include examples by default

Custom Skills Path

Override embedded skills with your customized versions:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"],
      "env": {
        "PM_SKILLS_PATH": "/path/to/my/custom/skills"
      }
    }
  }
}

Using Custom Skills from a pm-skills Fork

If you've forked pm-skills to customize skills:

Clone your fork locally
Make changes to skills in skills/{phase-skill}/ (e.g., skills/deliver-prd/)
Point pm-skills-mcp to your fork:

{
  "mcpServers": {
    "pm-skills": {
      "command": "npx",
      "args": ["pm-skills-mcp"],
      "env": {
        "PM_SKILLS_PATH": "/path/to/my-pm-skills-fork/skills"
      }
    }
  }
}

See the pm-skills authoring guide for skill modification guidelines.

(<a href="#readme-top">back to top</a>)

Usage

How It Works

┌─────────────────────────────────────────────────────────────┐
│                    Your AI Assistant                        │
│              (Claude, Cursor, Continue, etc.)               │
└──────────────────────────┬──────────────────────────────────┘
                           │ MCP Protocol
                           ▼
┌─────────────────────────────────────────────────────────────┐
│                    pm-skills-mcp Server                     │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│   ┌─────────────┐   ┌─────────────┐   ┌─────────────┐       │
│   │   59 Tools  │   │  Resources  │   │  3 Prompts  │       │
│   │             │   │             │   │             │       │
│   │ • 40 skills │   │ • templates │   │ • kickoff   │       │
│   │ • 11 flows  │   │ • examples  │   │ • lean      │       │
│   │ • 8 utils   │   │ • skills    │   │ • quick-prd │       │
│   │             │   │             │   │             │       │
│   └─────────────┘   └─────────────┘   └─────────────┘       │
│                                                             │
│   ┌─────────────────────────────────────────────────────┐   │
│   │              Embedded PM-Skills Library             │   │
│   │     40 skills × (SKILL.md + TEMPLATE + EXAMPLE)     │   │
│   └─────────────────────────────────────────────────────┘   │
│                                                             │
└─────────────────────────────────────────────────────────────┘

You ask your AI to create a PRD, hypothesis, or any PM artifact
AI invokes the appropriate pm_* tool via MCP
Server returns skill instructions, template, and optionally an example
AI generates a professional-quality artifact following the skill framework

Tools

PM-Skills MCP wraps each skill from pm-skills as an MCP tool. The 40 skill tools (like pm_prd, pm_hypothesis, pm_acceptance_criteria, pm_pm_skill_builder) generate PM artifacts, while 11 workflow tools and 8 utility tools help you orchestrate and validate skill usage. See the Comparison section for when to use MCP tools vs file-based slash commands.

Every skill tool accepts these parameters:

Parameter	Required	Description
`topic`	Yes	The subject or feature to create this artifact for. Be specific: "user authentication for mobile app" is better than "auth".
`context`	No	Additional requirements, constraints, or background. Use this for business context, technical constraints, or stakeholder needs.
`format`	No	Output verbosity: `full` (default, includes all guidance), `concise` (template + key points), or `template-only` (just structure).
`includeExample`	No	Set to `true` to include a completed example for reference. Helpful when learning a new skill.

Example invocations:

# Basic usage
Use pm_prd with topic "user authentication for mobile app"

# With context
Use pm_hypothesis with topic "checkout abandonment" and context "We see 40%
cart abandonment. Hypothesis: users abandon due to required account creation."

# Request concise output with example
Use pm_user_stories with topic "dark mode" format "concise" includeExample true

The Skills

🔍 Discover Phase - Find the right problem

Tool	Description
`pm_interview_synthesis`	Turn user research into actionable insights
`pm_competitive_analysis`	Map the landscape, find opportunities
`pm_stakeholder_summary`	Understand who matters and what they need

📋 Define Phase - Frame the problem

Tool	Description
`pm_problem_statement`	Crystal-clear problem framing
`pm_hypothesis`	Testable assumptions with success metrics
`pm_opportunity_tree`	Teresa Torres-style outcome mapping
`pm_jtbd_canvas`	Jobs to be Done framework

💡 Develop Phase - Explore solutions

Tool	Description
`pm_solution_brief`	One-page solution pitch
`pm_spike_summary`	Document technical explorations
`pm_adr`	Architecture Decision Records
`pm_design_rationale`	Why you made that design choice

🚀 Deliver Phase - Ship it

Tool	Description
`pm_prd`	Comprehensive product requirements
`pm_user_stories`	INVEST-compliant stories with acceptance criteria
`pm_edge_cases`	Error states, boundaries, recovery paths
`pm_launch_checklist`	Never miss a launch step again
`pm_release_notes`	User-facing release communication

📊 Measure Phase - Validate with data

Tool	Description
`pm_experiment_design`	Rigorous A/B test planning
`pm_instrumentation_spec`	Event tracking requirements
`pm_dashboard_requirements`	Analytics dashboard specs
`pm_experiment_results`	Document learnings from experiments

🔄 Iterate Phase - Learn and improve

Tool	Description
`pm_retrospective`	Team retros that drive action
`pm_lessons_log`	Build organizational memory
`pm_refinement_notes`	Capture backlog refinement outcomes
`pm_pivot_decision`	Evidence-based pivot/persevere framework

🧭 Foundation - Cross-cutting capability

Tool	Description
`pm_persona`	Generate product or marketing personas with explicit assumptions and evidence

🔗 Workflows - Multi-skill workflows

Tool	Effort	Description
`pm_workflow_feature_kickoff`	standard	problem → hypothesis → solution → PRD → stories
`pm_workflow_lean_startup`	comprehensive	hypothesis → experiment → results → pivot decision
`pm_workflow_triple_diamond`	comprehensive	Full discovery to delivery sequence
`pm_workflow_quick_prd`	quick	Fast problem → PRD workflow
`pm_workflow_experiment_cycle`	standard	hypothesis → experiment → results → lessons
`pm_workflow_customer_discovery`	standard	research → JTBD → opportunities → problem statement
`pm_workflow_sprint_planning`	quick	refinement → stories → edge cases
`pm_workflow_product_strategy`	comprehensive	competitive → stakeholders → opportunities → solution → ADR
`pm_workflow_post_launch_learning`	comprehensive	instrumentation → dashboard → results → retro → lessons
`pm_workflow_stakeholder_alignment`	standard	stakeholders → problem → solution → launch readiness
`pm_workflow_technical_discovery`	standard	spike → ADR → design rationale

🛠️ Utility Tools

Tool	Description
`pm_list_skills`	List all available PM skill tools
`pm_list_resources`	List all available MCP resources
`pm_list_workflows`	List all workflows with steps
`pm_list_prompts`	List available conversation prompts
`pm_validate`	Validate artifact against skill template
`pm_search_skills`	Search skills by keyword across names, descriptions, and content
`pm_cache_stats`	Show skill-cache hit rate, size, and TTL diagnostics

(<a href="#readme-top">back to top</a>)

MCP Resources

Access skill content directly via MCP resources:

pm-skills://skills/{skill}      → Full skill instructions
pm-skills://templates/{skill}   → Template only
pm-skills://examples/{skill}    → Worked example

Examples:

pm-skills://skills/deliver-prd
pm-skills://templates/define-hypothesis
pm-skills://examples/measure-experiment-design

Note: Phase information is available in resource metadata.

(<a href="#readme-top">back to top</a>)

Prompts

MCP prompts provide guided conversation starters for common workflows. Use prompts/get with the prompt name and a topic to begin.

Prompt	Description
`feature-kickoff`	Complete feature kickoff: Problem → Hypothesis → Solution → PRD → Stories
`lean-validation`	Build-Measure-Learn cycle: Hypothesis → Experiment → Results → Pivot
`quick-prd`	Fast PRD creation when requirements are clear: Problem → PRD

Usage example:

prompts/get name="feature-kickoff" arguments={"topic": "dark mode support"}

(<a href="#readme-top">back to top</a>)

Guides

Comprehensive documentation for setup, customization, and understanding pm-skills-mcp.

Guide	Description
Getting Started	Complete walkthrough from installation to your first tool invocation. Covers core concepts, platform-specific setup, workflows, and troubleshooting.
Integration Guide	Detailed setup instructions for each MCP client (Claude Desktop, Cursor, Continue, Cline, etc.) with troubleshooting tips.
Customization Guide	How to use custom skills with pm-skills-mcp. Covers environment variable overrides, forking pm-skills, and creating new skills.
Migration Guide	Moving between file-based pm-skills and MCP-based pm-skills-mcp. Includes command-to-tool mapping and hybrid approaches.
Architecture	Technical deep-dive for contributors. Covers server internals, data flow, extension points, and performance considerations.
Repository Governance	Top-level ownership and tracked-vs-local persistence policy for contributors.

(<a href="#readme-top">back to top</a>)

Project Status

Development Status

Maintenance mode as of 2026-05-04. Active feature development is paused pending demonstrated demand. The project will continue to receive security patches and critical bug fixes; new skill parity with the upstream pm-skills library is on hold pending a sustained signal of multi-team adoption.

Resumption criteria. A sustained signal of demand via GitHub Discussion engagement, install-volume growth, or direct contact from teams adopting the server in production workflows.

Recommended path for new users. The file-based install via the main pm-skills repository remains under active maintenance and is the supported path going forward. The main repository carries current parity with the latest pm-skills releases.

To register interest: open a GitHub Discussion.

Releases

All releases are available on the GitHub Releases page.

Install the latest version:

npm install -g pm-skills-mcp

Install a pinned release:

npm install -g pm-skills-mcp@2.9.3

From v2.4.0 onward, pm-skills-mcp directly tracks pm-skills release versions. Pinned source compatibility metadata is declared in pm-skills-source.json for each release.

Latest: v2.9.3 (security patch; 2026-05-05; catalog frozen at v2.9.2 build under maintenance mode). See GitHub Releases for the v2.9.x line. | Previous release details | Full changelog

Project Structure

See docs/reference/project-structure.md for detailed descriptions.

pm-skills-mcp/
├── src/                      # TypeScript source code
│   ├── index.ts              # Entry point
│   ├── server.ts             # MCP server implementation
│   ├── config.ts             # Configuration management
│   ├── cache.ts              # Skill caching layer
│   ├── skills/               # Skill loader and parser
│   ├── tools/                # MCP tool handlers (59 tools)
│   ├── resources/            # MCP resource handlers (skills/templates/examples + optional personas)
│   ├── prompts/              # MCP prompt definitions (3 prompts)
│   ├── workflows/            # Workflow definitions
│   └── types/                # TypeScript type definitions
├── skills/                   # Embedded PM skills (flat, copied from pm-skills v2.x)
│   ├── deliver-prd/          # Example: phase-prefixed skill directories
│   ├── define-hypothesis/    # Each skill has SKILL.md + references/
│   ├── discover-interview-synthesis/
│   └── ...                   # 40 skills total
├── docs/                     # Documentation
│   ├── getting-started.md    # Complete setup and first-use guide
│   ├── integration-guide.md  # Client-specific configuration
│   ├── customization.md      # Using custom skills with MCP
│   ├── migration-guide.md    # Moving between file-based and MCP
│   ├── architecture.md       # Technical internals for contributors
│   ├── releases/             # Shipped release notes
│   └── reference/            # Reference documentation
│       ├── project-structure.md
│       └── repository-governance.md
├── tests/                    # Vitest test suites (81 tests)
├── scripts/                  # Build and utility scripts
├── dist/                     # Compiled JavaScript output
├── .github/                  # CI/CD workflows
│   └── workflows/            # ci.yml, publish.yml, codeql.yml
├── pm-skills-source.json     # Pinned pm-skills repo/ref/version metadata
├── AGENTS.md                 # Agent discovery file
├── CONTRIBUTING.md           # Contribution guidelines
├── SECURITY.md               # Security policy
└── CHANGELOG.md              # Version history

Previous Release Details

<details> <summary>v2.6.0 - Maintenance: pm-skills v2.6.0 version parity</summary>

Version and source-pin metadata aligned with pm-skills v2.6.0.
No MCP tool/resource/prompt behavior changes.
Release note: docs/releases/Release_v2.6.0.md.

</details> <details> <summary>v2.5.x - Persona tool support + taxonomy contract updates</summary>

v2.5.2 - Public release-doc readability cleanup. v2.5.1 - Canonical AGENTS/claude continuity path. v2.5.0 - Persona skill tool (pm_persona), two-axis classification model (phase + classification), embed validation hardening. Tool count: 38.

Release notes: Release_v2.5.0.md through Release_v2.5.2.md.

</details> <details> <summary>v2.4.x - Direct version tracking with pm-skills</summary>

v2.4.3 - Release metadata/link alignment patch. v2.4.2 - Governance + structure-doc alignment. v2.4.1 - Version/pin parity patch. v2.4.0 - Adopted direct version tracking with pm-skills. Added pm-skills-source.json for reproducible embeds. Resource URI contract tests. Tool count: 36.

Release notes: Release_v2.4.0.md through Release_v2.4.3.md.

</details> <details> <summary>v2.1.0 and earlier</summary>

v2.1.0 - Flat skill structure alignment with pm-skills v2.x. Resource URIs flattened. v1.1.0 - Comprehensive documentation suite, platform compatibility. v1.0.0 - First stable release: 36 tools, caching, community governance. v0.1.x - Initial implementation: MCP server, CI/CD, npm packaging.

See CHANGELOG.md for full detail.

</details>

Changelog

See CHANGELOG.md for full version history.

Version	Date	Highlights
2.9.3	2026-05-05	Security patch: Dependabot alert cleanup (8 alerts closed). Catalog frozen at v2.9.2 build.
2.9.2	2026-05-04	Maintenance-mode declaration. 40 skills embedded; final new-skill embedding under active development.
2.9.0	2026-04-06	Workflow-tool parity with pm-skills (11 workflows), 59 tools total
2.8.0	2026-04-03	Lifecycle skill embedding (Create, Validate, Iterate)
2.7.0	2026-03-22	2 new skill tools (`pm_acceptance_criteria`, `pm_pm_skill_builder`), 27 skills, 40 tools
2.6.0	2026-03-04	Maintenance: pm-skills v2.6.0 version/source-pin parity
2.5.0	2026-03-02	Persona tool + taxonomy contract updates + embed hardening
2.4.0	2026-02-16	Direct version tracking + pinned source metadata
2.1.0	2026-01-27	Flat skill structure alignment with pm-skills v2.x
1.0.0	2026-01-21	First stable release: 36 tools, caching, governance
0.1.0	2026-01-20	Initial MCP server implementation

Roadmap

See the open issues for planned features.

[x] Core MCP server with all 40 PM skills
[x] Workflow tools (11 workflows)
[x] MCP resources for direct skill access
[x] MCP prompts for guided workflows
[x] Automated npm publishing
[x] Skill versioning and compatibility tracking
[ ] Additional workflows
[ ] Custom skill contribution support

(<a href="#readme-top">back to top</a>)

Development

Testing

The project uses Vitest for testing with 81 tests across multiple test files.

# Run all tests
npm test

# Watch mode for development
npm run test:watch

# Generate coverage report
npm run test:coverage

Test coverage includes:

Skill loader and parser
Tool handler and response formatting
Workflow execution
Prompt registration
Output validation
Server initialization

Code Quality

# Lint with ESLint
npm run lint

# Auto-fix lint issues
npm run lint:fix

# Format with Prettier
npm run format

# Check formatting
npm run format:check

# Type check
npx tsc --noEmit

CI/CD

The project uses GitHub Actions for continuous integration and deployment:

CI Workflow (.github/workflows/ci.yml)

Triggers on push/PR to main
Tests on Node.js 18, 20, and 22
Runs lint, format check, type check, and tests
Clones pm-skills at the pinned ref from pm-skills-source.json and embeds skills for testing

Publish Workflow (.github/workflows/publish.yml)

Triggers on GitHub release creation
Auto-publishes to npm with provenance
Uses npm automation token for authentication
Embeds skills from the same pinned pm-skills ref declared in pm-skills-source.json

CodeQL Workflow (.github/workflows/codeql.yml)

Security scanning for JavaScript/TypeScript
Runs on push to main and weekly schedule

(<a href="#readme-top">back to top</a>)

Security

We take security seriously. This project includes:

CodeQL Analysis - Automated security scanning on every push
Dependabot - Automated dependency updates for npm and GitHub Actions
Vulnerability Reporting - See SECURITY.md for our security policy

To report a vulnerability, please email security concerns privately rather than opening a public issue. See SECURITY.md for details.

(<a href="#readme-top">back to top</a>)

Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make will benefit everybody else and are greatly appreciated.

How to Contribute

Quick contribution steps:

Fork the Project
Create your Feature Branch (git checkout -b feature/AmazingFeature)
Commit your Changes using Conventional Commits (git commit -m 'feat: add amazing feature')
Push to the Branch (git push origin feature/AmazingFeature)
Open a Pull Request

Please read our CONTRIBUTING.md for detailed guidelines.

Reporting Bugs

Please try to create bug reports that are:

Reproducible - Include steps to reproduce the problem
Specific - Include as much detail as possible (version, environment, etc.)
Unique - Do not duplicate existing opened issues
Scoped - One bug per report

(<a href="#readme-top">back to top</a>)

About

Author

Howdy, I'm Jonathan Prisant, a product leader/manager/nerd in the church technology space who gets unreasonably excited about understanding + solving problems, serving humans, designing elegant systems, and getting stuff done. I enjoy optimizing and scaling workflows more than is probably healthy... NOT because I'm particularly fond of "business process definition", but because I think in systems and value the outcomes of increased "effectiveness and efficiency" (i.e. doing less of the boring work and more of the work I actually enjoy).

I am a follower of Jesus Christ, grateful husband to my beloved, proud (and exhausted) dad of 4 humans of various sizes and ages, D&D geek, 3d printing enthusiast, formerly-consistent strength trainer, smart home enthusiast, insatiable learner, compulsive tech-experimenter, writer-of-words that aggregate into sentences and paragraphs, and a bunch of other stuff too. I have too many projects going on across too many domains and need better self control, but hopefully you find this open-source repo helpful and useful.

If PM-Skills has helped you ship better products, consider giving the repo a star and sharing it with your team.

License

Distributed under the Apache License 2.0. See LICENSE for more information.

This means you can:

Use PM-Skills MCP commercially
Modify and distribute
Use privately
Include in proprietary software

The only requirements are attribution and including the license notice.

Security

We take security seriously. This project includes:

CodeQL Analysis - Automated security scanning on every push
Dependabot - Automated dependency updates for npm and GitHub Actions
Vulnerability Reporting - See SECURITY.md for our security policy

To report a vulnerability, please email security concerns privately rather than opening a public issue.

Acknowledgments

PM-Skills - The skill library that powers this MCP server. All 40 PM skills, templates, and examples come from pm-skills (embedded catalog frozen at the v2.9.2 build under maintenance mode).
Model Context Protocol - The protocol that makes this possible
Anthropic - For creating MCP and Claude

(<a href="#readme-top">back to top</a>)

Community

Have ideas for making PM-Skills MCP even better? Here are some ways to contribute and connect:

Feature Ideas

Open a feature request to suggest improvements
Join the Discussions to brainstorm with the community

Spread the Word

Give the repo a star if you find it useful
Share PM-Skills MCP with your team
Write about how you use PM-Skills MCP in your workflow

Feedback

Found something confusing? Open an issue
Want to chat? Start a discussion

Built with purpose by <a href="https://github.com/product-on-purpose">Product on Purpose</a> Bringing professional PM skills to every AI assistant

(<a href="#readme-top">back to top</a>)

Recommended Servers

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official

Featured

TypeScript

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official

Featured

Local

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official

Featured

TypeScript

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official

Featured

Python

E2B

Using MCP to run code via e2b.

Official

Featured

Neon Database

MCP server for interacting with Neon Management API and databases

Official

Featured

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official

Featured

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official

Featured

PM-Skills MCP

README

The Big Idea

Why MCP?

The Transformation

Key Features

Built with...

Works for...

Platform Compatibility

Comparison: pm-skills-mcp (this repo) vs. pm-skills

Getting Started

NPM Global Install

From Source

Quick Start by Platform

Configuration

Environment Variables

Custom Skills Path

Using Custom Skills from a pm-skills Fork

Usage

How It Works

Tools

The Skills

🔍 Discover Phase - Find the right problem

📋 Define Phase - Frame the problem

💡 Develop Phase - Explore solutions

🚀 Deliver Phase - Ship it

📊 Measure Phase - Validate with data

🔄 Iterate Phase - Learn and improve

🧭 Foundation - Cross-cutting capability

🔗 Workflows - Multi-skill workflows

🛠️ Utility Tools

MCP Resources

Prompts

Guides

Project Status

Development Status

Releases

Project Structure

Previous Release Details

Changelog

Roadmap

Development

Testing

Code Quality

CI/CD

Security

Contributing

How to Contribute

Reporting Bugs

About

Author

License

Security

Acknowledgments

Community

Recommended Servers

Comparison: `pm-skills-mcp` (this repo) vs. `pm-skills`