mcp-camunda
Tools to interact with the Camunda 7 Community Edition Engine using the Model Context Protocol (MCP). Whether you're automating workflows, querying process instances, or integrating with external systems, Camunda MCP Server is your agentic solution for seamless interaction with Camunda.
claude mcp add --transport stdio lepoco-mcp-camunda docker run -d -i --rm --name mcp-camunda mcp/camunda
How to use
This MCP server provides a bridge to interact with a Camunda 7 Community Engine using the Model Context Protocol (MCP). It exposes a set of tools to query and manage Camunda data, including listing deployments, instances, variables, incidents, and user tasks, as well as counting and resolving certain entities. You can run commands against the running container to perform actions like listing definitions, listing or counting process instances, and resolving incidents. Use the available tools to automate workflows or integrate Camunda processes with other systems using MCP requests and responses supported by this server.
To start using the server, deploy the container as shown in the installation guide and ensure your Camunda Engine is reachable (e.g., via CAMUNDA_HOST in HTTP mode if needed). Interact with the MCP tools by issuing MCP requests to the server’s endpoints or through the MCP client you’re using, following the standard MCP command/response patterns for the listed tools such as list_definitions, list_instances, count_variables, list_user_tasks, and resolve_incident.
How to install
Prerequisites:
- Docker installed and running on your host
- Access to a Camunda 7 Community Edition Engine (or a compatible Camunda REST endpoint)
Installation steps:
-
Pull and run the MCP Camunda image in detached mode:
docker run -d -i --rm --name mcp-camunda mcp/camunda
This launches the MCP Camunda server using the default configuration. If you need HTTP mode or a specific Camunda host, you can pass environment variables as shown in the examples (e.g., -e MODE=Http -e CAMUNDA_HOST=http://host.docker.internal:8080/engine-rest/).
-
Verify the server is running:
docker ps | grep mcp-camunda
-
(Optional) Create or edit your .mcp.json configuration to point to this server if you’re using an external MCP client:
{ "servers": { "camunda": { "type": "stdio", "command": "docker", "args": [ "run", "-d", "-i", "--rm", "mcp/camunda" ] } }, "inputs": [] }
-
Start using the available MCP tools (see how_to_use for details).
Additional notes
Environment variables and options you may encounter:
- MODE: Set to Http to enable HTTP mode (see README examples).
- CAMUNDA_HOST: The base URL to connect to the Camunda Engine REST API (e.g., http://host.docker.internal:8080/engine-rest/).
- If you run in HTTP mode, you may need to expose and map ports as shown in the README (e.g., -p 64623:8080).
Common issues:
- Ensure the Camunda Engine is reachable from the MCP container; network or host address misconfigurations are typical sources of failure.
- When using docker, the container name in MCP config should match or align with the actual container you’re running. Adjust the --name flag as needed.
- If you see authentication or CORS errors from Camunda, verify REST API accessibility and any required security settings on your Camunda instance.
Related MCP Servers
sandbox
A Model Context Protocol (MCP) server that enables LLMs to run ANY code safely in isolated Docker containers.
mcp-playground
A Streamlit-based chat app for LLMs with plug-and-play tool support via Model Context Protocol (MCP), powered by LangChain, LangGraph, and Docker.
better-notion
Markdown-first MCP server for Notion API - composite tools optimized for AI agents
nmap
MCP server for AI-powered network scanning with Nmap. Port scanning, service detection, OS fingerprinting, and vulnerability scanning for AI agents. By Vorota AI.
mcp-demo
Example of using MCP Gateway with E2B sandboxes
SqlAugur
MCP server providing AI assistants with safe, read-only access to SQL Server databases. AST-based query validation, rate limiting, and DBA diagnostic tooling.