MCP Client
Overview
Mendix MCP Client Module
Consume external tools and prompts from your enterprise landscape within your Mendix app using the Mendix MCP Client module.
The Mendix MCP Client Module provides an easy low-code capability to set up a connection to an MCP Server from a Mendix app using the Model Context Protocol (MCP). Using the module, resources such as tools and prompts can easily be discovered and used within a Mendix application, without any custom integrations to third-party systems. The Mendix MCP Client module builds a bridge between Mendix and MCP server applications such as other Mendix apps, through the MCP Java SDK.
With the current implementation, it is possible to:
- Discover prompts and tools from servers.
- Consume reusable prompts including the ability to use prompt arguments
- Call external tools as part of an LLM interaction (if combined with GenAI Commons)
Check out the GenAI Showcase App for an example implementation.
Contact Us
For support and questions, feel free to reach out via email: genai-components-feedback@mendix.com or Community Slack.
Documentation
Mendix MCP Client Module
How to use
For an example implementation check out the GenAI Showcase App.
Known Limitations
- Tools and prompt messages can only return String content.
Releases
- We upgraded the MCP Java SDK version to 1.1.1.
- We no longer append the endpoint with /mcp or /sse, so that MCP servers can be consumed that don't follow this pattern.
Note: This means that if you connected to MCP servers before, you need to ensure that the endpoint is set correctly with /mcp or /sse (if applicable).
This is a replacement of the earlier released module 3.0.0, which has been unpublished.
User Access and Approval is a new GenAI Commons and ConversationalUI feature to make tool calls visible to users and provide control for users to decide if tools should be executed or rejected. They can be either configured at runtime in AgentCommons or at design time using GenAI Commons actions when adding tools/functions to a request. Per default all tools are hidden for users (same behavior as in previous versions). You can optionally provide a display title and description to be shown to the user. When adding MCP tools to a request, you can now specify those fields as well. If a whole server gets added, you can only configure the User Access and Approval for all tools in general.
The module has been aligned with the latest changes in ConsumedKnowledgeBase and ConsumedMCPService. This is a breaking change and might need migration to prevent loss of data and functionality. For migration steps, please refer to the migration guide here: https://docs.mendix.com/appstore/modules/genai/genai-for-mx/migration-guide/
This module version is compatible with the GenAI Commons 6.0.0 version.
The module now supports more complex tools to be called, for example with array parameters or more descriptive schemas. For this MCP tools now contain the exact schema that the MCP server exposed and pass the exact tool input from the model to the MCP server.
The module is now compatible with the latest version of GenAI Commons.. Make sure to also update applicable LLM connectors.
* We added support for Streamable HTTP transport method which can be enabled when using the 2025-03-26 protocol version during creation of the client connection.
* The Create MCP Client action now throws an error when it cannot connect to the server. This can be used to implement custom error handling (for example to change the authentication headers). We recommend to add error handling in all places where this action is implemented.
* We added more debug logging to the List Tools and List Prompts actions, to understand what the server returned.
* We added ConnectionTimeOutInSeconds to the MCPServerConfiguration entity to change the timeout settings. This can happen either in the UI via the config page or when creating the connection in a microflow. The default timeout is 10 seconds.
* The module was upgraded to 10.24.13.
Small internal change for compatibility with new ConversationalUI v5.3.0 capability: Observability for AI Traces
Static image widget instances were replaced with Image widgets.
MCP tool sync process has been improved. Sync process is added to task queue and completed in the background