Deployment

Deployment Options

gitlab-api exposes its MCP server (console script gitlab-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. Replace the <your-…> placeholders with the values from the Configuration / Environment Variables 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": {
    "gitlab-mcp": {
      "command": "uvx",
      "args": ["--from", "gitlab-api", "gitlab-mcp"],
      "env": {
        "GITLAB_URL": "<your-gitlab_url>"
      }
    }
  }
}

2. Streamable-HTTP (local process)

Run the server as a long-lived HTTP process:

uvx --from gitlab-api gitlab-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": {
    "gitlab-mcp": {
      "command": "uvx",
      "args": ["--from", "gitlab-api", "gitlab-mcp", "--transport", "streamable-http", "--port", "8000"],
      "env": {
        "TRANSPORT": "streamable-http",
        "HOST": "0.0.0.0",
        "PORT": "8000",
        "GITLAB_URL": "<your-gitlab_url>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "gitlab-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": {
    "gitlab-mcp": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "TRANSPORT=stdio",
        "-e", "GITLAB_URL=<your-gitlab_url>",
        "knucklessg1/gitlab-api:latest"
      ]
    }
  }
}

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

docker run -d --name gitlab-mcp -p 8000:8000 \
  -e TRANSPORT=streamable-http \
  -e PORT=8000 \
  -e GITLAB_URL="<your-gitlab_url>" \
  knucklessg1/gitlab-api:latest
# or, from a clone of this repo:
docker compose -f docker/mcp.compose.yml up -d

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

(c) From a local checkout with uv:

uv run gitlab-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": {
    "gitlab-mcp": { "url": "http://gitlab-mcp.arpa/mcp" }
  }
}

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

This page covers running gitlab-api as long-lived servers: the transports, a Docker Compose stack, the optional A2A agent server, putting it behind a Caddy reverse proxy, and giving it a DNS name with Technitium. To provision the GitLab instance it connects to, see Backing Platform.

gitlab-api ships both an MCP server (console script gitlab-mcp) and an A2A agent server (console script gitlab-agent). The MCP server is a typed, deterministic tool surface a policy router / agent calls; the agent server is a Pydantic-AI graph agent that wraps that surface with an AG-UI web interface.

Run the MCP server

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

stdio (default)streamable-httpsse

gitlab-mcp

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

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

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

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

Health check (HTTP transports):

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

Configuration (environment)

gitlab-api is configured entirely from the environment. The required set:

Var	Default	Meaning
GITLAB_URL	https://gitlab.com	GitLab instance base URL
GITLAB_TOKEN	(unset)	Personal / project / group access token
GITLAB_SSL_VERIFY	True	Verify TLS (set False for self-signed homelab)
HOST	0.0.0.0	Bind address (HTTP transports)
PORT	8000	Listen port (HTTP transports)
TRANSPORT	stdio	stdio, streamable-http, or sse

Each tool module is independently togglable with a *TOOL flag (for example BRANCHESTOOL, PIPELINESTOOL, GRAPHQLTOOL) — every variable, with its default, is documented in .env.example. Copy it to .env and fill in only what you use; the connector remains inactive when GITLAB_TOKEN is absent.

Docker Compose

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

services:
  gitlab-api-mcp:
    image: knucklessg1/gitlab-api:latest
    container_name: gitlab-api-mcp
    hostname: gitlab-api-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 GITLAB_URL / GITLAB_TOKEN
docker compose -f docker/mcp.compose.yml up -d
docker compose -f docker/mcp.compose.yml logs -f

Agent server (A2A)

gitlab-api also ships an A2A agent server (console script gitlab-agent). It runs the Pydantic-AI graph agent, calls the MCP tool surface over MCP_URL, and exposes an AG-UI web interface on port 9017. The repo ships docker/agent.compose.yml, which deploys the MCP server and the agent server together:

services:
  gitlab-api-mcp:
    image: knucklessg1/gitlab-api:latest
    container_name: gitlab-api-mcp
    hostname: gitlab-api-mcp
    restart: always
    env_file:
      - ../.env
    environment:
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
    ports:
      - "8000:8000"

  gitlab-api-agent:
    image: knucklessg1/gitlab-api:latest
    container_name: gitlab-api-agent
    hostname: gitlab-api-agent
    restart: always
    depends_on:
      - gitlab-api-mcp
    env_file:
      - ../.env
    command: ["gitlab-agent"]
    environment:
      - HOST=0.0.0.0
      - PORT=9017
      - MCP_URL=http://gitlab-api-mcp:8000/mcp
      - PROVIDER=${PROVIDER:-openai}
      - MODEL_ID=${MODEL_ID:-gpt-4o}
      - ENABLE_WEB_UI=True
    ports:
      - "9017:9017"

docker compose -f docker/agent.compose.yml up -d
curl -s http://localhost:9017/health        # {"status":"OK"}

The agent reaches the MCP server by container name through MCP_URL; set PROVIDER and MODEL_ID (plus the matching provider API key) to select the model.

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
gitlab-api.arpa {
    tls internal
    reverse_proxy gitlab-api-mcp:8000
}

# Public — automatic Let's Encrypt
gitlab-api.example.com {
    reverse_proxy gitlab-api-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=gitlab-api.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 gitlab-api.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:

{
  "mcpServers": {
    "gitlab-api": {
      "command": "uv",
      "args": ["run", "gitlab-mcp"],
      "env": {
        "GITLAB_URL": "https://gitlab.example.com",
        "GITLAB_TOKEN": "<your-gitlab-token>"
      }
    }
  }
}

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