Developing MCP tools

Expose your IBM® App Connect Enterprise REST APIs as MCP (Model Context Protocol) tools that AI agents can discover and use to interact with your integration solutions.

About this task

The Model Context Protocol (MCP) is an open-source protocol that enables AI agents to discover and interact with your IBM App Connect Enterprise integration solutions. By exposing your REST APIs as MCP tools, you can make your integrations available to AI-powered applications and agents without requiring code changes to your existing flows.

MCP support is available in IBM App Connect Enterprise v13.0.7 or later. This feature enables you to:
  • Expose REST API flows as discoverable tools for AI agents
  • Enable AI agents to automatically understand and use your integration capabilities
  • Provide standardized access to your integrations through the MCP protocol
  • Configure and manage MCP server settings through the integration server configuration

To develop and deploy MCP tools in IBM App Connect Enterprise, you need to understand the MCP protocol, configure your integration server to expose REST APIs as MCP tools, and test your tools with AI agents or MCP clients.

Procedure

Read the following sections to learn about developing MCP tools in IBM App Connect Enterprise:

  • What is MCP?: Learn about the Model Context Protocol, how it works, and how IBM App Connect Enterprise implements MCP server functionality to expose your REST APIs as tools for AI agents.
  • Exposing a REST API as an MCP tool: Follow step-by-step instructions to expose a REST API as an MCP tool, including configuring the MCP server and deploying your REST API.
  • Configuring MCP properties: Reference documentation for all MCP configuration properties in the server.conf.yaml file, including startup modes, port configuration, authentication, and TLS settings.

What to do next

After exposing your REST APIs as MCP tools, you can:
  • Integrate your MCP tools with AI platforms and applications that support the Model Context Protocol
  • Monitor MCP server activity and tool usage through integration server logs
  • Configure security settings such as authentication and TLS to protect your MCP endpoints
  • Expose additional REST APIs as MCP tools to expand the capabilities available to AI agents

For more information about creating and deploying REST APIs in App Connect Enterprise, see Creating a REST API and Packaging and deploying a REST API.