Anyone in SF? We'd love to meet you tomorrow 👉 https://lu.ma/fqdql7kh 🚀 Tomorrow, come and hack with us to build the MCP discovery system using Elastic and mcp-use (YC S25)! The challenge is to build a better discovery layer: an index, a search interface, a ranking system, something an agent could actually use to choose which server or tool to invoke! Who will join?

We are live on Product Hunt! 🚀 Thanks Garry Tan for hunting us! Check it out 👉 https://lnkd.in/erCCCDdu We are launching our cloud platform! With our cloud we allow you to host, deploy, test, mix and match MCP servers to create super powerful agents with access to your internal services and thousands of other applications!

mcp-use (YC S25) is live on product hunt, we are launching our cloud platform! 🔥 With our cloud we allow you to host, deploy, test, mix and match MCP servers to create super powerful agents with access to your internal services and thousands of others applications. This is me and Luigi Pederzani at midnight following the launch, support us at https://lnkd.in/dSJx5vfM 🤗

Next Tuesday, mcp-use (YC S25) and Elastic will co host an MCP Hack Night 🚀 We'll deep dive into an interesting question: How will AI agents discover and search for MCPs and how will that differ from how humans search for information on the internet today? The Model Context Protocol (MCP) team is working on a official registry where all the MCP servers will be listed, though it is still unclear how agents will be able to take advantage of that autonomously. Elastic has been powering large-scale search for over a decade, in products used by millions, helping enterprises make sense of vast amounts of data in real time. The same capability that agents will need to search thousands of MCP servers. Expect an evening of hands on exploration, great conversations, most importantly building. If you want to contribute to the future of the web for agents, do not miss. RSVP link in the comments!

We at mcp-use are developing the open-source dev tools and infrastructure for MCP to help dev teams quickly build and deploy custom AI agents with MCP servers. Our SDK just crossed 100,000 downloads and 4,800 Github stars. Dev teams use us at a variety of companies, including both startups and enterprises such as NASA, NVIDIA, SAP… Building an agent with or without MCP? Let's chat

mcp-use (YC S25) and Langfuse integration docs have just been released! You can monitor and trace your MCP Agent interactions for enhanced observability. On our way to becoming the go-to SDK for MCP-enabled AI agents.  👇 Link in the comments!

mcp-use (YC S25) python library just crossed 100,000 downloads! 🚀 Back to work 🫡

Want to stand out as an SWE when top open-source startups open a new role? The secret is simple: start contributing early! If you're interested in getting involved with one of the most popular MCP libraries out there, hop into our Discord. (We've got an exclusive channel just for contributors 👀)👇

mcp-use helps developers easily build, deploy, and manage MCP agents without worrying about infrastructure. We solve this by providing an SDK and a unified platform vertical for MCP development, enabling teams to focus entirely on their core products. The SDK is the most popular MCP library with: - 4,000+ Github stars - 80,000+ downloads - 480+ forks It's used by devs at global companies including NVIDIA, Red Hat, Oracle, IBM, Intuit, Elastic, and Verizon. Let's chat if you need help with MCP servers or agents development! 👇