About Cumulocity MCP Server
Cumulocity MCP Server exposes device, measurement, and alarm functions from the Cumulocity IoT platform to MCP clients. It also evaluates JSONata expressions against JSON objects.
Key features
- 01Filter devices by type or name
- 02Retrieve device details, children, and fragments
- 03Get measurements within a date range
- 04Query active alarms by severity
- 05Evaluate JSONata expressions on JSON
Use cases
- 01Find devices by type or name
- 02Inspect device measurements over a time period
- 03Filter and monitor active alarms
- 04Apply JSONata to JSON data
MCP support
The server exposes Cumulocity device management, measurement, alarm, and JSONata evaluation tools to MCP clients. It can run locally or be deployed as a Cumulocity microservice.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMECumulocity MCP Server
A Python-based server that provides Cumulocity IoT platform functionality through the MCP (Model Control Protocol) interface. This server enables seamless interaction with Cumulocity's device management, measurements, and alarm systems.
Available Tools
Device Management
-
Get Devices
- List and filter devices
- Parameters:
type: Filter by device typename: Filter by device namepage_size: Results per page (max 2000)current_page: Page number
-
Get Device by ID
- Retrieve detailed information for a specific device
- Parameter:
device_id: Device identifier
-
Get Child Devices
- View child devices of a specific device
- Parameter:
device_id: Parent device identifier
-
Get Device Fragments
- Access device fragments and their values
- Parameter:
device_id: Device identifier
Measurements
Get Device Measurements
- Retrieve device measurements with time filtering
- Parameters:
device_id: Device identifierdate_from: Start date (ISO 8601 format)date_to: End date (ISO 8601 format)page_size: Number of measurements to retrieve
Alarms
Get Active Alarms
- Monitor active alarms in the system
- Parameters:
severity: Filter by severity levelpage_size: Number of results to retrieve
Dynamic Mapper
evaluate_jsonata_expression Evaluate a JSONata expression against a given JSON object.
Input: A JSON object as string and a JSONata expression string. Output: Result of the JSONata expression evaluation.
Installation & Deployment
Local Installation
Using uv (recommended)
When using uv no specific installation is needed for this package. We will
use uvx to directly run mcp-server-c8y.
Using PIP
Alternatively you can install mcp-server-c8y via pip:
pip install mcp-server-c8y
After installation, you can run it as a script using:
python -m mcp_server_c8y
Deployment on Cumulocity Tenant
You can deploy this server as a Cumulocity microservice for direct integration with your tenant. This is done by uploading a special deployment package (mcp-server-c8y.zip) to your Cumulocity tenant.
Building the Microservice Deployment Package
- Ensure you have Docker and zip installed on your system.
- Run the provided build script to create the deployment package:
./scripts/buildcontainer.sh
This will:
- Build the Docker image for the microservice
- Save the image as
image.tarin thedocker/directory - Package
image.tarandcumulocity.jsonintodocker/mcp-server-c8y.zip
Deploying to Cumulocity
- Log in to your Cumulocity tenant as a user with microservice deployment permissions.
- Navigate to Administration > Ecosystem > Microservices.
- Click Add microservice and upload the
mcp-server-c8y.zipfile from thedocker/directory. - Wait for the microservice to be deployed and started. You should see its status as "Available" once ready.
- The microservice will be accessible under your tenant's service URL, typically:
https://<your-tenant>.cumulocity.com/service/mcp-server-c8y/mcp/
For more details on Cumulocity microservice deployment, see the official documentation.
Usage with Claude Desktop
This MCP Server can be used with Claude Desktop to enable Claude to interact with your Cumulocity IoT platform. Follow these steps to set it up:
-
Download and install Claude Desktop
-
Configure Claude Desktop to use this MCP Server:
- Open Claude Desktop
- Click on the Claude menu and select "Settings..."
- Navigate to "Developer" in the left-hand bar
- Click "Edit Config"
-
Add the following configuration to your
claude_desktop_config.json:
Using uvx
"mcpServers": {
"mcp-c8y": {
"command": "uvx",
"args": [
"mcp-server-c8y",
"--transport",
"stdio"
],
"env": {
"C8Y_BASEURL": "https://your-cumulocity-instance.com",
"C8Y_TENANT": "your-tenant-id",
"C8Y_USER": "<your-username>",
"C8Y_PASSWORD": "<your-password>"
}
}
}
Replace the following placeholders with your actual values:
https://your-cumulocity-instance.com: Your Cumulocity instance URLyour-tenant-id: Your Cumulocity tenant IDyour-username: Your Cumulocity usernameyour-password: Your Cumulocity password
-
Restart Claude Desktop
-
You should now see a hammer icon in the bottom right corner of the input box. Click it to see the available Cumulocity tools.
For more detailed information about using MCP Servers with Claude Desktop, visit the official MCP documentation.
Cursor MCP Server Settings Example
If you are using Cursor and have deployed your MCP Server to a Cumulocity tenant, you can configure your MCP server connection with a .cursor/mcp.json file. Example (with sensitive data anonymized):
{
"mcpServers": {
"Cumulocity": {
"url": "https://your-cumulocity-instance.com/service/mcp-server-c8y/mcp/",
"headers": {
"Authorization": "Basic <YOUR_BASE64_AUTH_TOKEN>"
}
}
}
}
https://your-cumulocity-instance.com: Your Cumulocity instance URL- Replace
<YOUR_BASE64_AUTH_TOKEN>with your actual Base64-encoded credentials. Never commit real credentials to version control.
Contributing
We welcome contributions from everyone! Here's how you can contribute to this project:
- Fork the repository
- Create a new branch for your feature or bugfix
- Make your changes following these best practices:
- Write clear, descriptive commit messages
- Follow the existing code style and conventions
- Add tests for new features
- Update documentation as needed
- Ensure all tests pass
- Submit a pull request
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.