Product Hunt MCP
Connect Product Hunt data to any LLM or agent
143 followers
Connect Product Hunt data to any LLM or agent
143 followers
Connect Product Hunt's API to LLMs through the Model Context Protocol. Access posts, collections, topics, users, votes, and comments with Claude Desktop, Cursor, or any MCP client.
Free
Launch Team / Built With
Render Sandboxes: a computer for every agentGive every agent its own machine. Starts in under a second
Promoted
👋 Hey Product Hunt!
I was surprised to find that there was no MCP for Product Hunt yet, despite the GraphQL API being available. So I built one!
This server connects Product Hunt's API to any LLM that speaks the Model Context Protocol, letting your AI assistants access real-time Product Hunt data without complex coding.
With this MCP, your AI can answer questions like
- What were the top 5 launches last December?
- What's the difference between average vote counts in December 2023 vs December 2024?
- Which topics had the most popular products this quarter?
Key features
- Get info on posts, comments, collections, topics, users
- Search/filter by topic, date, votes
- Built with FastMCP for speed
It's perfect for AI enthusiasts using Claude Desktop or Cursor, developers building with Product Hunt data, or anyone exploring the MCP ecosystem.
The repo has setup instructions. Would love your feedback! 🚀
@chrismessina At least for the APIs that I tried (getting posts with different parameter values). Any specifics that I should be checking?
@jaipandya it's definitely not working for me.
What redirect_uri should I be using for my Product Hunt app?
I did follow your instructions and set everything up, but clearly I'm getting errors:
@chrismessina Thanks for the feedback, Chris. For redirect_uri, you can use a localhost URL like "https://localhost:8742/". We don't use it in the app though, because we use developer token to call the PH API.
Your issue is probably Claude specific. For some reason, it doesn't have access to your PATH variables it seems. So you might need to give it the full path to your executable.
Once you have installed the server using `pip install product-hunt-mcp`. You can get the executable path by using `which product-hunt-mcp`. In your Claude MCP config, put the following:
Quit and reopen Claude again after making these changes. And it should show the 11 tools the MCP server makes available to you.
Does it connect to the PH api?
Will it tell me all the people that voted for a product?
@kensavage It does use PH API behind the scene. PH API response is limited in terms of user information though (because of privacy concerns). It's good with count of votes, comments posted on a product launch etc, but when it comes to enquiring about users, it doesn't help you much.
Very cool, @jaipandya! I think there's a lot of opportunity to integrate this along with other product discovery/review sites and social platforms to get a better signal on what's trending and/or most useful.
@rrhoover Thanks so much, Ryan! And I agree a hundred percent. PH carries strong signal for what is trending right now. I found MCP provided context specially helpful while comparing different time segments for product trends, and that tell you a lot of about how things are changing, and where you should be focusing. For example comparing top products from March '24 vs March '25.
What was your primary motivation to create this?
@busmark_w_nika Thanks for asking! I’ve always found Product Hunt to be a goldmine of emerging trends, ideas, and community sentiment. But getting meaningful insights out of it usually meant wrangling APIs or relying on someone technical.
With the Model Context Protocol (MCP), that barrier disappears. It democratizes access—suddenly, anyone can start asking rich, exploratory questions like:
-What kind of products are trending this week in AI?
-How have indie hacker tools evolved over the past year?
-Which categories get the most community engagement?
And then this goldmine opens up to product makers, with more insights and data you can take better decisions.
@jaipandya I find it helpful, as together with some guys are trying to pick non-featured products. There are many good ones, and I even bought them. This can be a huge help! :)
@sahithkrishna_an Thank you, Sahith!
Direct API integration makes life so much easier, thank you! 👍
@shenjun Thank you, Jun. I'm happy that you found it useful.