mcp-use is the open-source devtools and cloud infrastructure to help dev teams quickly build and deploy custom AI agents with MCP servers. Our SDK has over 5,000 GitHub stars, 100k downloads, and is trusted by engineers at NASA, Cisco, NVIDIA, etc...
MRKTPLACE
Congrats on the launch! Excited to see the MCP ecosystem evolve further :)
mcp-use
@rishabhjain1198 Thank you! :)
Congrats on the launch! Another big fan of MCP here
Fantastic work, love how the product is evolving, keep it up 🚀
Happy to support an amazing project and stellar team
Congrats! Time to fly high and make an impact! Gooo 🇮🇹
VibeFlow
Congrats guys! Quick one on observability: Can I see per-tool traces with full request/response and agent decisions, and filter by server or profile?
Aryel
Super exciting to see a vertical stack for MCP. The SDK + control plane combo is exactly what teams need to move from local experiments to production—especially with safer defaults (profiles, audit logs) and observability built in. Curious to see more on server pooling and sandboxed stdio in real-world deployments. Congrats on the launch!
mcp-use
@leonardo_ubbiali Thanks! That’s exactly the gap we set out to close.
With mcp-use, the SDK is the quick start for local agents.
Then the control plane brings the production-grade pieces:
MCP servers orchestration,
MCP agents runtine handled by us,
centralized config,
profile-based access control,
observability
So teams aren’t stitching these together by hand.