Build AI Agents with MCP

Databases
Use MCP in Jet Admin
Overview

The MCP integration connects Jet Admin to any Model Context Protocol server, so you can query a server's resources, list its available tools, and trigger a tool call directly from a Jet Admin app. It's a way to give an AI agent standardized access to your own data and workflows.

Once connected, MCP tool calls can be scoped by role, joined to records in your own database, and gated behind an approval step before they run. Every call is recorded in Jet Admin's audit log alongside the rest of your app's activity.

What you can do
Query an MCP server
Call a connected MCP server's resources directly from a Jet Admin action.
List available tools
List the tools an MCP server exposes so you know what's callable.
Trigger a tool call
Invoke a specific MCP tool call and return its result to your app.
Use Cases
Scenario
What it looks like
Give an AI agent access to your data
Let an AI agent read and act on your own database through a standard MCP connection.
Centralize tool access for AI workflows
Expose a consistent set of tools to any MCP-compatible AI client.
Debug and inspect MCP tool calls
See exactly which MCP tools were called and what they returned, for troubleshooting.

Key benefits

  • Standardized AI access. Let an AI agent read and act on your data through a common protocol.
  • Centralized tool access. Expose one consistent set of tools to any MCP-compatible client.
  • Debuggable by design. Inspect exactly which tools were called and what they returned.
  • Full audit trail. Every MCP tool call made from Jet Admin is logged for review.

How to connect Jet Admin with MCP

  1. In Jet Admin, select MCP as a Data Source.
  2. Enter the MCP server's connection details.
  3. List the tools it exposes and choose which to use.
  4. Test the connection to confirm tool calls run correctly.
  5. Build a table, form, or automated action using MCP tool calls.