PM-Copilot MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@PM-Copilot MCP Generate a PRD for a mobile checkout feature"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
PM-Copilot MCP
An AI-powered Product Management Operating System exposed as a Model Context Protocol server. It gives Product Managers, founders, and product leaders a set of tools, prompts, and reference resources for the workflows they do every day — PRDs, user stories, prioritization, competitor analysis, roadmaps, KPIs, meeting notes, and product strategy — directly from Claude Desktop, Claude Code, Cursor, Windsurf, or any other MCP-compatible client.
How generation works
PM-Copilot does not call out to an LLM API itself. Tools that produce narrative documents (PRDs,
user stories, competitor analysis, roadmaps, KPIs, meeting summaries, strategy docs) use the MCP
sampling capability to ask the connected client's own model to do the writing, using a
PM-best-practices system prompt and a structured request. prioritize_features is fully
deterministic: it computes RICE and ICE scores and MoSCoW buckets in code.
Related MCP server: Tycana
Tools
Tool | Purpose |
| Enterprise-grade Product Requirement Document |
| INVEST-compliant user stories with acceptance criteria and edge cases |
| RICE / ICE / MoSCoW scoring and final prioritization |
| Feature comparison matrix, SWOT, and strategic recommendations |
| Quarterly roadmap with milestones, dependencies, risks |
| North Star metric with leading/lagging indicators |
| Summary, decisions, risks, action items, follow-ups |
| Vision, mission, goals, market analysis, positioning, growth plan |
Resources
pm-copilot://templates/{prd,roadmap,user-story,strategy}— reusable document templatespm-copilot://frameworks— reference definitions for RICE, ICE, MoSCoW, Kano, JTBD, Opportunity Solution Tree, SWOT, Porter's Five Forces, and PESTLE
Prompts
create_prd, generate_user_stories_prompt, analyze_competitor, build_product_strategy,
create_quarterly_roadmap — each wraps the corresponding tool with the right arguments.
Project structure
src/
tools/ one module per MCP tool
resources/ templates + framework library, registered as MCP resources
prompts/ guided prompt wrappers around the tools
schemas/ Zod input schemas and shared validation helpers
utils/ structured logger and the sampling helper
server.ts builds and wires the McpServer
cli.ts stdio entrypoint (`pm-copilot-mcp` binary)
tests/ Jest unit testsDevelopment
npm install
npm run build # compile TypeScript to dist/
npm test # run the Jest test suite
npm run start # run the compiled server over stdioDocker
docker build -t pm-copilot-mcp .
docker run -i pm-copilot-mcpThe server communicates over stdio (the standard MCP transport for desktop clients), so run the
container interactively (-i) rather than publishing a port.
Claude Desktop configuration
{
"mcpServers": {
"pm-copilot": {
"command": "node",
"args": ["/path/to/pm-copilot-mcp/dist/cli.js"]
}
}
}Quality
All tool inputs are validated with Zod; invalid input raises a descriptive error before any generation happens.
Structured JSON logs are written to stderr so they never collide with MCP JSON-RPC traffic on stdout.
Tool results return both human-readable text and
structuredContentfor programmatic clients.
Roadmap
Planned premium features include PM intelligence (usage/retention analytics), Jira/Notion/Linear integrations, and an AI product coach for interview and case-study practice — see the original product brief for the full commercialization plan.
Available Tools
8 toolsanalyze_meeting_notesAnalyze Meeting NotesB
Extract a summary, decisions, risks, action items, and follow-ups from meeting notes.
Inputs: meeting_notes (required).
competitor_analysisCompetitor AnalysisC
Analyze competitors and identify strategic opportunities using a SWOT-style breakdown.
Inputs: product_name (required), competitors (required).
generate_kpisGenerate KPIsC
Generate a KPI framework with a North Star Metric, leading/lagging indicators, and dashboards.
Inputs: product_type (required), goal (required).
generate_prdGenerate PRDC
Create an enterprise-grade Product Requirement Document from a feature idea.
Inputs: product_name (required), feature_name (required), problem_statement (required), target_users (required), +1 more.
generate_roadmapGenerate RoadmapC
Generate a quarterly roadmap with milestones, dependencies, risks, and success indicators.
Inputs: product_vision (required), features (required).
generate_strategyGenerate StrategyC
Build a product strategy document: vision, mission, goals, market analysis, positioning, growth plan.
Inputs: market (required), product (required), business_goal (required).
generate_user_storiesGenerate User StoriesC
Generate INVEST-compliant agile user stories with acceptance criteria and edge cases.
Inputs: feature_description (required).
prioritize_featuresPrioritize FeaturesA
Rank features using RICE, ICE, and MoSCoW frameworks. Accepts plain feature names or objects with reach/impact/confidence/effort/ease to refine the scoring; missing fields use sensible defaults.
Inputs: features (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 8 tools
Each tool targets a distinct product management task: meeting analysis, competitor analysis, KPI generation, PRD creation, roadmap, strategy, user stories, and prioritization. There is no functional overlap.
Most tools follow a verb_noun pattern (generate_*, analyze_meeting_notes, prioritize_features), but 'competitor_analysis' deviates to noun_noun. Overall consistent and readable.
With 8 tools, the set is well-scoped for a product management assistant, covering essential planning and analysis tasks without unnecessary bloat.
The tools cover core planning and analysis functions (PRD, roadmap, strategy, user stories, prioritization, KPIs, competitor analysis). Missing operational features like tracking or updating items, but comprehensive for the stated purpose.
Maintenance
Related MCP Connectors
AI-powered product management: backlog optimization, scheduling, forecasting, and PRD generation.
Your product team's shared strategic memory — an MCP server your AI tools reason over.
Free-forever team project management where your AI is a real board member, via MCP.
AIOProductOS: revenue, feedback, work and analytics on one customer record. 71 tools, 3 apps.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceTransforms scattered customer feedback from sources like Slack, Zoom, and JIRA into actionable product insights and AI-generated PRDs. It features over 50 tools for semantic clustering, sentiment analysis, and VOC-based prioritization to streamline product management workflows.1MIT
- AlicenseNot gradedqualityDmaintenanceAI-native productivity backend that gives your AI assistant persistent memory, pattern awareness, and computed intelligence about your work. 14 MCP tools for task management, daily planning, weekly review, and personal context.MIT
- AlicenseAqualityDmaintenanceMCP server exposing 40 best-practice product management skills plus workflows and lifecycle tools, instantly accessible to any AI via Model Context Protocol.19152 npm21Apache 2.0
- FlicenseNot gradedqualityBmaintenanceA single source of truth for a product's state, roadmap, and deferred work, read and written equally by you and your AI agent via web dashboard and MCP.-