Xano MCP Servers have arrived!
MCP (Model Context Protocol) is an emerging standard that enables AI agents to interact with external services in a consistent, structured way. Rather than building custom scripts or one-off integrations every time an agent needs to act beyond its core capabilities, you can now connect it to an MCP Server to handle the task.
With Xano, you can now build and host your own MCP Servers using the same powerful visual builder you already use for APIs and logic. This unlocks a new way to design agent-compatible backends—going beyond simple endpoints and building for AI, right inside Xano.
Perform one-off updates to table content seamlessly without creating a workflow
Xano’s MCP Server enables users to run sophisticated CRM queries, like this one from a sales ops analyst.
Power AI-driven database updates on your marketplace app backend with tools hosted on your own MCP Server built in Xano
If you aren't familiar with MCP, or maybe this is the first time you've heard of it: that's okay, a lot of us were in the same boat. It's easy to get up to speed right inside of the Xano docs. We've built a library of information on key concepts, how to get started, and an end-to-end example of Xano MCP servers in action.
Your backend can now act as an MCP Client
Xano has introduced two new intrinsic functions that allow your backend to interact with external MCP Servers:
MCP List Tools – retrieve a list of available Tools from a remote MCP Server
MCP Call Tool – execute a specific Tool with input parameters
These functions can be used directly within API endpoints, custom functions, or MCP Server Tools—no special configuration required.
With these additions, your backend can dynamically discover and invoke tools hosted on other MCP Servers, enabling powerful AI workflows that extend beyond your local environment. Whether you're supporting chatbot behavior or coordinating multi-step processes, these new client capabilities help you build smarter, more connected systems with full control.
Swagger Docs Upgrade - Define Custom Inputs and Responses
When you build API endpoints in Xano, they're already auto-documented using the standard OpenAPI specification (powered by Swagger), making it easier for you to pass API documentation to users and other members of your team.
We've introduced the ability for you to define sample inputs and responses, greatly increasing the usefulness of your API documentation, especially when passing it to an AI model for development assistance.
You can quickly add them after running the API inside of Xano, or modify them directly from your API settings at any time.
AI Database Assistant Upgrades
You can now delete tables and columns using the assistant.
You're now able to generate default CRUD endpoints when creating new tables. It will ask you if you'd like to add them to an existing API group, or create a new one, and you'll also be able to enable authentication on all of the new endpoints in one swing.
