概要
What is ProductPlan MCP Server?
An MCP server that lets AI assistants (Claude, Cursor, and others) query and manage ProductPlan roadmaps, OKRs, ideas, and launches through natural conversation. It connects directly to the ProductPlan API and runs as a local subprocess, keeping your API token on your machine.
How to use ProductPlan MCP Server?
Download the correct binary for your OS from the releases page, place it in a known path (e.g., /usr/local/bin/productplan on Mac/Linux), then configure your AI assistant’s MCP settings with the binary path and your ProductPlan API token. After setup, you can ask natural language questions like “What’s on our Q1 roadmap?” or “Create a new idea for mobile app improvements.”
Key features of ProductPlan MCP Server
- Read, create, update, and delete roadmaps, bars, milestones, ideas, objectives, and more
- Works with Claude Desktop, Claude Code, Cursor, VS Code (Cline, Continue), and n8n
- Pre-built agent skills for product managers, executives, and customer-facing roles
- No coding required – configure with a JSON file and paste your API token
- Local process keeps your token secure; no third-party servers involved
- Real‑time synchronous communication via stdin/stdout
Use cases of ProductPlan MCP Server
- Morning standup prep: summarize recent roadmap changes in a week
- Stakeholder updates: list all Q1 objectives and their progress
- Idea triage: find ideas tagged “enterprise” that lack a priority
- Launch coordination: check incomplete tasks for an upcoming launch
- Quick lookups: ask when a specific bar is scheduled to start
FAQ from ProductPlan MCP Server
Which AI assistants are supported?
Claude Desktop, Claude Code (terminal), Cursor, VS Code with the Cline or Continue extensions, and n8n workflow automation.
Do I need to write any code?
No – you only need to copy a file, paste an API token, and follow the setup instructions. No programming required.
Where is my API token stored?
Your API token is set as an environment variable in your AI assistant’s MCP configuration. It never passes through third‑party servers and stays on your local machine.
What ProductPlan data can I access?
Roadmaps (view), bars (full CRUD), bar comments, connections, links, lanes, milestones, ideas, idea customers/tags, opportunities, forms, objectives, key results, and launches (view only). See the data access table in the README for details.
What are “agent skills”?
Agent skills are pre‑built workflow guides that train your AI assistant how to use ProductPlan tools effectively for specific personas: general product management, leadership, and customer‑facing roles. They can be copied into your AI assistant’s skills directory.
