Overview
What is Product Hunt MCP Server?
A plug-and-play MCP server that connects Product Hunt’s API to any LLM or agent supporting the Model Context Protocol, enabling access to Product Hunt posts, comments, collections, topics, and user data.
How to use Product Hunt MCP Server?
Install with pip install product-hunt-mcp, set the PRODUCT_HUNT_TOKEN environment variable with your Product Hunt API developer token, and run product-hunt-mcp. For Claude Desktop or Cursor, add a JSON configuration block specifying the command and environment variable.
Key features of Product Hunt MCP Server
- Get details on posts, comments, collections, topics, users
- Search and filter by topic, date, votes, and more
- Paginated results for comments and user upvotes
- Built with FastMCP for performance and MCP compatibility
- Docker support for containerized deployment
Use cases of Product Hunt MCP Server
- AI assistants that fetch live Product Hunt data for users
- Bots or dashboards that monitor trending products and discussions
- Automations that collect product feedback or analyze community topics
FAQ from Product Hunt MCP Server
What do I need to run this server?
Python 3.10+ and a Product Hunt API developer token (obtain one from your Product Hunt API dashboard).
How do I authenticate the server?
Set the PRODUCT_HUNT_TOKEN environment variable to your developer token. The token is required and must be passed in your MCP client’s environment configuration.
Where does the data come from?
All data is fetched in real‑time from the official Product Hunt API. The server does not store or cache data locally.
What transport does the server use?
The server communicates via standard MCP stdio transport. It runs as a command‑line process that your MCP client launches (e.g., via command: "product-hunt-mcp" in Claude Desktop config).
Are there any rate limits?
Yes, the Product Hunt API enforces rate limits. The client respects these limits and will report when they reset if exceeded. You can check your current limit with the check_server_status or get_api_rate_limits tools.
