Deployment¶
Deployment Options¶
mattermost-mcp exposes its MCP server (console script mattermost-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": {
"mattermost-mcp": {
"command": "uvx",
"args": ["--from", "mattermost-mcp", "mattermost-mcp"],
"env": {
"MATTERMOST_URL": "<your-mattermost_url>",
"MATTERMOST_TOKEN": "<your-mattermost_token>"
}
}
}
}
2. Streamable-HTTP (local process)¶
Run the server as a long-lived HTTP process:
uvx --from mattermost-mcp mattermost-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": {
"mattermost-mcp": {
"command": "uvx",
"args": ["--from", "mattermost-mcp", "mattermost-mcp", "--transport", "streamable-http", "--port", "8000"],
"env": {
"TRANSPORT": "streamable-http",
"HOST": "0.0.0.0",
"PORT": "8000",
"MATTERMOST_URL": "<your-mattermost_url>",
"MATTERMOST_TOKEN": "<your-mattermost_token>"
}
}
}
}
…or connect to the already-running process by URL:
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": {
"mattermost-mcp": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"-e", "TRANSPORT=stdio",
"-e", "MATTERMOST_URL=<your-mattermost_url>",
"-e", "MATTERMOST_TOKEN=<your-mattermost_token>",
"knucklessg1/mattermost-mcp:latest"
]
}
}
}
(b) Run a local streamable-http container, then connect by URL:
docker run -d --name mattermost-mcp -p 8000:8000 \
-e TRANSPORT=streamable-http \
-e PORT=8000 \
-e MATTERMOST_URL="<your-mattermost_url>" \
-e MATTERMOST_TOKEN="<your-mattermost_token>" \
knucklessg1/mattermost-mcp:latest
# or, from a clone of this repo:
docker compose -f docker/mcp.compose.yml up -d
(c) From a local checkout with uv:
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:
Caddy reverse-proxies http://mattermost-mcp.arpa to the container's :8000
streamable-http listener; http://mattermost-mcp.arpa/health returns
{"status":"OK"} when the service is live.
This page covers running mattermost-mcp as a long-lived server: the transports, a
Docker Compose stack, the companion agent server, putting it behind a Caddy reverse
proxy, and giving it a DNS name with Technitium. To provision the Mattermost server
it connects to, see Backing Platform.
mattermost-mcpships both an MCP server (console scriptmattermost-mcp) and a Pydantic-AI agent server (console scriptmattermost-agent). The MCP server is a typed, deterministic tool surface a policy router / agent calls; the agent server consumes that tool surface and can serve a conversational web UI.
Run the MCP server¶
The transport is selected with --transport (or the TRANSPORT env var):
Health check (HTTP transports):
Configuration (environment)¶
mattermost-mcp is configured entirely from the environment. The required set:
| Var | Default | Meaning |
|---|---|---|
MATTERMOST_URL |
http://localhost:8065 |
Mattermost Server v4 API endpoint URL |
MATTERMOST_TOKEN |
(none) | Personal Access Token or Bot Token |
Plus HOST / PORT / TRANSPORT for HTTP transports. The full template, ready to
copy to .env, is documented in
.env.example.
The connector remains inactive when MATTERMOST_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:
mattermost-mcp:
image: knucklessg1/mattermost-mcp:latest
container_name: mattermost-mcp
hostname: mattermost-mcp
restart: always
env_file:
- ../.env
environment:
- PYTHONUNBUFFERED=1
- HOST=0.0.0.0
- PORT=8000
- TRANSPORT=streamable-http
- MATTERMOST_URL
- MATTERMOST_TOKEN
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 MATTERMOST_* values
docker compose -f docker/mcp.compose.yml up -d
docker compose -f docker/mcp.compose.yml logs -f
Agent server¶
mattermost-mcp also ships a Pydantic-AI agent (console script mattermost-agent)
that connects to the running MCP server over HTTP and optionally serves a web UI. The
repo provides docker/agent.compose.yml,
which runs the MCP server and the agent together: the agent reaches the MCP server by
container name via MCP_URL and publishes its own web UI on :9035.
services:
mattermost-mcp-mcp:
image: knucklessg1/mattermost-mcp:latest
container_name: mattermost-mcp-mcp
hostname: mattermost-mcp-mcp
restart: always
env_file:
- ../.env
environment:
- PYTHONUNBUFFERED=1
- HOST=0.0.0.0
- PORT=8000
- TRANSPORT=streamable-http
ports:
- "8000:8000"
mattermost-mcp-agent:
image: knucklessg1/mattermost-mcp:latest
container_name: mattermost-mcp-agent
hostname: mattermost-mcp-agent
restart: always
depends_on:
- mattermost-mcp-mcp
env_file:
- ../.env
command: [ "mattermost-agent" ]
environment:
- PYTHONUNBUFFERED=1
- HOST=0.0.0.0
- PORT=9035
- MCP_URL=http://mattermost-mcp-mcp:8000/mcp
- PROVIDER=${PROVIDER:-openai}
- MODEL_ID=${MODEL_ID:-gpt-4o}
- ENABLE_WEB_UI=True
ports:
- "9035:9035"
docker compose -f docker/agent.compose.yml up -d
curl -s http://localhost:9035/health # agent health
Run the agent directly against an existing MCP server:
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
mattermost-mcp.arpa {
tls internal
reverse_proxy mattermost-mcp:8000
}
Reload Caddy:
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=mattermost-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 mattermost-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 mm):
{
"mcpServers": {
"mattermost-mcp": {
"command": "uv",
"args": ["run", "mattermost-mcp"],
"env": {
"MATTERMOST_URL": "http://your-mattermost:8065",
"MATTERMOST_TOKEN": "your_personal_access_token"
}
}
}
}
For a remote HTTP server, point the client at http://mattermost-mcp.arpa/mcp instead.