Deployment

Deployment Options

archimate-mcp exposes its MCP server (console script archimate-mcp) four ways. Pick the row that matches where the server runs relative to your MCP client, then copy the matching mcp_config.json below. Add the service-connection environment variables documented in the Configuration section.

#	Option	Transport	Where it runs	mcp_config.json key
1	stdio	stdio	client launches a subprocess	command
2	Streamable-HTTP (local)	streamable-http	a local network port	command or url
3	Local container / uv	stdio or streamable-http	Docker / Podman / uv on this host	command or url
4	Remote URL	streamable-http	a remote host behind Caddy	url

1. stdio (local subprocess)

The client launches the server over stdio via uvx — best for local IDEs (Cursor, Claude Desktop, VS Code):

{
  "mcpServers": {
    "archimate-mcp": {
      "command": "uvx",
      "args": ["--from", "archimate-mcp", "archimate-mcp"]
    }
  }
}

2. Streamable-HTTP (local process)

Run the server as a long-lived HTTP process:

uvx --from archimate-mcp archimate-mcp --transport streamable-http --host 0.0.0.0 --port 8000
curl -s http://localhost:8000/health        # {"status":"OK"}

Then either let the client launch it:

{
  "mcpServers": {
    "archimate-mcp": {
      "command": "uvx",
      "args": ["--from", "archimate-mcp", "archimate-mcp", "--transport", "streamable-http", "--port", "8000"],
      "env": {
        "TRANSPORT": "streamable-http",
        "HOST": "0.0.0.0",
        "PORT": "8000"
      }
    }
  }
}

…or connect to the already-running process by URL:

{
  "mcpServers": {
    "archimate-mcp": { "url": "http://localhost:8000/mcp" }
  }
}

3. Local container / uv

(a) Launch a container directly from mcp_config.json (stdio over the container — no ports to manage). Swap docker for podman for a daemonless runtime:

{
  "mcpServers": {
    "archimate-mcp": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "TRANSPORT=stdio",
        "knucklessg1/archimate-mcp:latest"
      ]
    }
  }
}

(b) Run a local streamable-http container, then connect by URL:

docker run -d --name archimate-mcp -p 8000:8000 \
  -e TRANSPORT=streamable-http \
  -e PORT=8000 \
  knucklessg1/archimate-mcp:latest
# or, from a clone of this repo:
docker compose -f docker/mcp.compose.yml up -d

{
  "mcpServers": {
    "archimate-mcp": { "url": "http://localhost:8000/mcp" }
  }
}

(c) From a local checkout with uv:

uv run archimate-mcp --transport streamable-http --port 8000

4. Remote URL (deployed behind Caddy)

When the server is deployed remotely (e.g. as a Docker service) and published through Caddy on the internal *.arpa zone, connect with the "url" key — no local process or image required:

{
  "mcpServers": {
    "archimate-mcp": { "url": "http://archimate-mcp.arpa/mcp" }
  }
}

Caddy reverse-proxies http://archimate-mcp.arpa to the container's :8000 streamable-http listener; http://archimate-mcp.arpa/health returns {"status":"OK"} when the service is live.

This page covers running archimate-mcp as a long-lived server: the transports, a Docker Compose stack, the companion A2A agent server, putting it behind a Caddy reverse proxy, and giving it a DNS name with Technitium.

archimate-mcp ships two entry points: an MCP server (console script archimate-mcp) and an A2A agent server (console script archimate-agent). The MCP server is a typed, deterministic tool surface a policy router or agent calls; the agent server wraps that surface as a Pydantic-AI agent for graph-orchestrated workflows.

Run the MCP server

The transport is selected with --transport (or the TRANSPORT env var):

stdio (default)streamable-httpsse

archimate-mcp

For IDE / desktop MCP clients that launch the server as a subprocess.

archimate-mcp --transport streamable-http --host 0.0.0.0 --port 8000

A network server with a /health endpoint and /mcp route.

archimate-mcp --transport sse --host 0.0.0.0 --port 8000

Health check (HTTP transports):

curl -s http://localhost:8000/health        # {"status":"OK"}

Configuration (environment)

archimate-mcp is configured entirely from the environment. The required set:

Var	Default	Meaning
ARCHI_MODEL_PATH	./model.archimate	Working model file (Open Exchange Format)
ARCHITOOL	True	Register the ArchiMate tool set

Plus HOST / PORT / TRANSPORT for HTTP transports. Copy .env.example to .env and adjust the values you need.

Docker Compose

The repo ships docker/mcp.compose.yml. A production stack reads a sibling .env, persists the model file on a volume, and publishes the HTTP server on :8000:

services:
  archimate-mcp:
    image: knucklessg1/archimate-mcp:latest
    container_name: archimate-mcp
    hostname: archimate-mcp
    restart: always
    env_file:
      - .env
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
      - ARCHI_MODEL_PATH=/data/model.archimate
    volumes:
      - archimate_data:/data
    ports:
      - "8000:8000"
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
      interval: 30s
      timeout: 10s
      retries: 3

volumes:
  archimate_data:

cp .env.example .env          # then edit ARCHI_MODEL_PATH if needed
docker compose -f docker/mcp.compose.yml up -d
docker compose -f docker/mcp.compose.yml logs -f

A2A agent server

The companion agent server (console script archimate-agent) wraps the same tool surface as a Pydantic-AI agent. It connects to a running MCP server via MCP_URL and exposes the agent capabilities declared in a2a.json.

# Point the agent at the MCP server and serve it on :8001
export MCP_URL=http://archimate-mcp:8000/mcp
archimate-agent --host 0.0.0.0 --port 8001

A combined stack from docker/agent.compose.yml runs the MCP server and wires the agent to it by container name:

services:
  archimate-mcp:
    image: knucklessg1/archimate-mcp:latest
    hostname: archimate-mcp
    environment:
      - TRANSPORT=streamable-http
      - HOST=0.0.0.0
      - PORT=8000
      - ARCHI_MODEL_PATH=/data/model.archimate
    volumes:
      - archimate_data:/data
    ports:
      - "8000:8000"

  archimate-agent:
    image: knucklessg1/archimate-mcp:latest
    command: ["archimate-agent", "--host", "0.0.0.0", "--port", "8001"]
    depends_on: [archimate-mcp]
    environment:
      - MCP_URL=http://archimate-mcp:8000/mcp
      - HOST=0.0.0.0
      - PORT=8001
    ports:
      - "8001:8001"

volumes:
  archimate_data:

docker compose -f docker/agent.compose.yml up -d

Behind a Caddy reverse proxy

Expose the HTTP server on a hostname with automatic TLS. Add to your Caddyfile:

# Internal (self-signed) — homelab .arpa zone
archimate-mcp.arpa {
    tls internal
    reverse_proxy archimate-mcp:8000
}

# Public — automatic Let's Encrypt
archimate-mcp.example.com {
    reverse_proxy archimate-mcp:8000
}

Reload Caddy:

docker compose -f services/caddy/compose.yml exec caddy caddy reload --config /etc/caddy/Caddyfile

DNS with Technitium

Point the hostname at the host running Caddy. Via the Technitium API:

curl -s "http://technitium.arpa:5380/api/zones/records/add" \
  --data-urlencode "token=$TECHNITIUM_DNS_TOKEN" \
  --data-urlencode "domain=archimate-mcp.arpa" \
  --data-urlencode "zone=arpa" \
  --data-urlencode "type=A" \
  --data-urlencode "ipAddress=10.0.0.10" \
  --data-urlencode "ttl=3600"

…or add an A record archimate-mcp.arpa → <caddy-host-ip> in the Technitium web console (http://technitium.arpa:5380). The ecosystem technitium-dns-mcp automates this as a tool.

Register with an MCP client

Add to your client's mcp_config.json (multiplexer nickname archi):

{
  "mcpServers": {
    "archimate-mcp": {
      "command": "uv",
      "args": ["run", "archimate-mcp"],
      "env": {
        "ARCHI_MODEL_PATH": "./model.archimate",
        "ARCHITOOL": "True"
      }
    }
  }
}

For a remote HTTP server, point the client at http://archimate-mcp.arpa/mcp instead.