Deployment

Deployment Options

wger-agent supports local stdio, a loopback-only development listener, a least-privilege stdio container, and a remote authenticated HTTPS boundary. Provider endpoint, credential, selector, identity, and trust material are supplied at runtime through AgentConfig; none is stored in this repository.

Installed stdio process

{
  "mcpServers": {
    "wger": {
      "command": "wger-mcp",
      "args": [],
      "env": {"MCP_TOOL_MODE": "intent"}
    }
  }
}

Loopback development listener

wger-mcp --transport streamable-http --host 127.0.0.1 --port 8000

Do not expose this listener beyond loopback. Network deployments require direct TLS or an explicitly trusted TLS-terminating ingress, configured authentication, exact MCP_ALLOWED_HOSTS, and an exact trusted-proxy CIDR policy.

Least-privilege local container

docker run -i --rm \
  --read-only \
  --cap-drop=ALL \
  --security-opt=no-new-privileges \
  --pids-limit=256 \
  --tmpfs /tmp:rw,noexec,nosuid,nodev,size=64m \
  -e TRANSPORT=stdio \
  registry.example.invalid/wger-agent@sha256:<digest> wger-mcp

The operator projects the selected AgentConfig profile into the process at runtime; the image remains immutable and contains no environment connection profile.

Remote authenticated HTTPS endpoint

{
  "mcpServers": {
    "wger": {"url": "https://service.example.invalid/mcp"}
  }
}

Store the real remote URL, outbound identity reference, and TLS-profile reference in AgentConfig, not in MCP client JSON or documentation.

This page covers running wger-agent as long-lived servers: the transports, a Docker Compose stack, the optional graph agent, putting it behind a Caddy reverse proxy, and giving it a DNS name with Technitium. To provision the Wger platform it connects to, see Backing Platform.

wger-agent ships two console scripts: an MCP server (wger-mcp) and a Pydantic-AI graph agent (wger-agent). The MCP server is a typed, deterministic tool surface; the agent server orchestrates those tools behind the Agent Control Protocol and the Agent Web UI.

Run the MCP server

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

stdio (default)streamable-httpsse

wger-mcp

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

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

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

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

Health check (HTTP transports):

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

Configuration (environment)

wger-agent is configured entirely from the environment. The required set:

Var	Default	Meaning
WGER_URL	https://wger.de	Wger instance base URL
WGER_API_KEY	(none)	Wger API token
WGER_DEFAULT_EMAIL	(none)	Default account email
WGER_DEFAULT_PASSWORD	(none)	Default account password
TRANSPORT	stdio	stdio, streamable-http, or sse
HOST	0.0.0.0	Bind address (HTTP transports)
PORT	8000	Bind port (HTTP transports)

Each tool domain is registered through a toggle variable — ROUTINETOOL, ROUTINECONFIGTOOL, EXERCISETOOL, WORKOUTTOOL, NUTRITIONTOOL, BODYTOOL, USERTOOL (all default True). Telemetry (ENABLE_OTEL, OTLP exporter settings) and access governance (EUNOMIA_TYPE, EUNOMIA_POLICY_FILE) are configured the same way. The full set is documented in .env.example. Copy it to .env and populate only what you use; the connector remains inactive when credentials are absent.

Docker Compose

The repo ships docker/mcp.compose.yml. It reads a sibling .env and publishes the HTTP server on :8000:

services:
  wger-agent-mcp:
    image: example/wger-agent@sha256:<digest>
    container_name: wger-agent-mcp
    hostname: wger-agent-mcp
    restart: always
    env_file:
      - ../.env
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
    ports:
      - "8000:8000"
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
      interval: 30s
      timeout: 10s
      retries: 3

cp .env.example .env          # then edit WGER_* values
docker compose -f docker/mcp.compose.yml up -d
docker compose -f docker/mcp.compose.yml logs -f

Agent server

To run the integrated Pydantic-AI graph agent, use the wger-agent console script. It connects to the MCP server over MCP_URL, exposes the Agent Control Protocol and the Agent Web UI on its own port (9004 by convention), and routes each request to the relevant tool domain.

export WGER_URL=https://your-wger:8000
export WGER_API_KEY=your_api_key
export MCP_URL=http://wger-agent-mcp:8000/mcp

wger-agent --provider openai --model-id gpt-4o

The repo ships docker/agent.compose.yml, which deploys the MCP server and the agent together on one network so the agent reaches the MCP server by container name:

services:
  wger-agent-mcp:
    image: example/wger-agent@sha256:<digest>
    container_name: wger-agent-mcp
    hostname: wger-agent-mcp
    restart: always
    env_file:
      - ../.env
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
    ports:
      - "8000:8000"

  wger-agent-agent:
    image: example/wger-agent@sha256:<digest>
    container_name: wger-agent-agent
    hostname: wger-agent-agent
    restart: always
    depends_on:
      - wger-agent-mcp
    env_file:
      - ../.env
    command: ["wger-agent"]
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=9004
      - MCP_URL=http://wger-agent-mcp:8000/mcp
      - PROVIDER=${PROVIDER:-openai}
      - MODEL_ID=${MODEL_ID:-gpt-4o}
      - ENABLE_WEB_UI=True
      - ENABLE_OTEL=True
    ports:
      - "9004:9004"

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 .example.invalid zone
wger-agent.example.invalid {
    tls internal
    reverse_proxy wger-agent-mcp:8000
}

# Public — automatic Let's Encrypt
wger-agent.example.com {
    reverse_proxy wger-agent-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.example.invalid:5380/api/zones/records/add" \
  --data-urlencode "token=$TECHNITIUM_DNS_TOKEN" \
  --data-urlencode "domain=wger-agent.example.invalid" \
  --data-urlencode "zone=arpa" \
  --data-urlencode "type=A" \
  --data-urlencode "ipAddress=192.0.2.10" \
  --data-urlencode "ttl=3600"

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

Register with an MCP client

Add to your client's mcp_config.json:

{
  "mcpServers": {
    "wger-agent": {
      "command": "uvx",
      "args": ["--from", "wger-agent", "wger-mcp"],
      "env": {
        "WGER_URL": "https://your-wger:8000",
        "WGER_API_KEY": "your_api_key"
      }
    }
  }
}

For a remote HTTP server, point the client at http://wger-agent.example.invalid/mcp instead.