Camunda Mcp

API | MCP Server | A2A Agent

Camunda (7 & 8) process automation API + MCP Server + A2A Agent for the agent-connector-sdk ecosystem.

Version: 0.5.0

Documentation — Installation, deployment, usage across the API, CLI, and MCP interfaces, and guidance for provisioning the Camunda platform are maintained in the official documentation.

camunda-mcp wraps the Camunda process automation APIs with thin, deterministic MCP tools, targeting both platforms from a single server: the Camunda 7 Engine REST API and the Camunda 8 Zeebe / Operate / Tasklist REST APIs. It also ships agent for agent-to-agent orchestration.

What it provides

• Api (camunda_mcp.api_client) — a facade holding both the Camunda 7 Engine REST client (Camunda7Api) and the Camunda 8 Zeebe / Operate / Tasklist client (Camunda8Api), each constructed lazily so a server configured for one platform never needs the other's URLs or credentials.
• Action-dispatch MCP tools (camunda-mcp console script): camunda_process, camunda_instance, camunda_task, camunda_deploy, camunda_message, camunda_external_task, camunda_job, camunda_history, camunda_deployment, and camunda_ops. Each takes a platform argument (7 or 8). See docs/overview.md for the full surface. a2a.json) wrapping the tool surface in a Pydantic-AI graph agent.

Available MCP Tools

Auto-generated — do not edit (synced by the mcp-readme-table pre-commit hook).

Intent tools

11 intent tool(s) · 0 operation(s). Call an operation as act(action='<operation>', params={...}) (ask for reads); find(action='describe', params={'action': '<operation>'}) returns its arguments. An operation is enabled unless its <DOMAIN>TOOL toggle is set false. Auto-generated — do not edit.

Environment Variables

Package environment variables

Inherited agent-connector-sdk variables (apply to every connector)

17 package + 23 inherited variable(s). Auto-generated from .env.example + the shared agent-connector-sdk set — do not edit.

Every variable the server reads, grouped by purpose. Credentials left blank leave the corresponding platform inactive — the connector remains inactive when credentials are absent. A starter .env.example ships with the repository; copy it to .env and populate the values for the platform the operator use.

Connection & Credentials

MCP server / transport

Tool toggles

Each action-routed tool can be disabled individually via its toggle env var (set to false). See the Available MCP Tools table above for the authoritative names.

Telemetry & governance

Agent CLI (full [agent] runtime only)

Installation

Pick the extra that matches what the operator want to run:

# Connector-focused MCP server (includes the shared graph engine)
uv pip install "camunda-mcp[mcp]"

# Agent runtime (adds model orchestration to the shared graph engine)

# Everything (development)
uv pip install "camunda-mcp[all]"      # or: python -m pip install "camunda-mcp[all]"

Run the servers:

camunda-mcp                       # stdio MCP server (default transport)
camunda-mcp --transport streamable-http --host 0.0.0.0 --port 8000

# A2A agent server against a deployed MCP endpoint
MCP_URL=http://camunda-mcp:8000/mcp camunda-agent --host 0.0.0.0 --port 8001

Container images (:mcp)

One multi-stage docker/Dockerfile builds the MCP-only image (--target mcp):

docker build --target mcp   -t example/camunda-mcp:mcp    docker/   # connector-focused MCP server

agent (immutable agent digest) with a co-located :mcp sidecar.

Knowledge-graph database (epistemic-graph)

Agent Connector SDK core dependency (epistemic-graph[full]). The [mcp] extra keeps deployments can use the bundled engine. For production or shared state, run epistemic-graph as a dedicated database service and configure the runtime to use it. Deployment recipes (single-node + Raft HA), connection configuration, and architecture diagrams are documented in the epistemic-graph deployment guide.

MCP config

Register in the multiplexer under nickname camun (tools surface as camun__process, camun__instance, camun__task, …). See camunda_mcp/mcp_config.json.

Additional Deployment Options

camunda-mcp can run as a local stdio process or container, or behind a remote network boundary. The Deployment guide carries the detailed transport contract.

• Local container — launch a reviewed immutable image as a least-privilege stdio child with no listener or published port.
• Remote URL — connect through an operator-supplied authenticated HTTPS ingress. Keep its URL, outbound identity references, trust profile, and exact MCP_ALLOWED_HOSTS in AgentConfig.

Documentation

The complete documentation is published as the official documentation site and is the recommended reference for installation, deployment, and day-to-day operation.

AGENTS.md is the canonical contributor/agent guidance.

Governed capability contract

This package ships a compact canonical skill surface with specialist procedures kept as referenced workflows. The current MCP tools, skill metadata, connector_manifest.yml, ontology, mappings, shapes, fixtures, migrations, tool-schema fingerprints, and certification metadata form one versioned capability contract. Validate them together; do not rely on stale tool names or historical per-task skill wrappers.

Runtime endpoints, credentials, certificate trust, tenant identity, retention, and observability policy are deployment inputs and are never packaged values. See Configuration, trust, and privacy before enabling a network transport, connector ingestion, Arbora delegation, or trace export.