Deployment¶
Deployment Options¶
archivebox-api exposes its MCP server (console script archivebox-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": {
"archivebox-mcp": {
"command": "uvx",
"args": ["--from", "archivebox-api", "archivebox-mcp"],
"env": {
"ARCHIVEBOX_BASE_URL": "<your-archivebox_base_url>",
"ARCHIVEBOX_URL": "<your-archivebox_url>",
"ARCHIVEBOX_USERNAME": "<your-archivebox_username>"
}
}
}
}
2. Streamable-HTTP (local process)¶
Run the server as a long-lived HTTP process:
uvx --from archivebox-api archivebox-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": {
"archivebox-mcp": {
"command": "uvx",
"args": ["--from", "archivebox-api", "archivebox-mcp", "--transport", "streamable-http", "--port", "8000"],
"env": {
"TRANSPORT": "streamable-http",
"HOST": "0.0.0.0",
"PORT": "8000",
"ARCHIVEBOX_BASE_URL": "<your-archivebox_base_url>",
"ARCHIVEBOX_URL": "<your-archivebox_url>",
"ARCHIVEBOX_USERNAME": "<your-archivebox_username>"
}
}
}
}
…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": {
"archivebox-mcp": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"-e", "TRANSPORT=stdio",
"-e", "ARCHIVEBOX_BASE_URL=<your-archivebox_base_url>",
"-e", "ARCHIVEBOX_URL=<your-archivebox_url>",
"-e", "ARCHIVEBOX_USERNAME=<your-archivebox_username>",
"knucklessg1/archivebox-api:latest"
]
}
}
}
(b) Run a local streamable-http container, then connect by URL:
docker run -d --name archivebox-mcp -p 8000:8000 \
-e TRANSPORT=streamable-http \
-e PORT=8000 \
-e ARCHIVEBOX_BASE_URL="<your-archivebox_base_url>" \
-e ARCHIVEBOX_URL="<your-archivebox_url>" \
-e ARCHIVEBOX_USERNAME="<your-archivebox_username>" \
knucklessg1/archivebox-api: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://archivebox-mcp.arpa to the container's :8000
streamable-http listener; http://archivebox-mcp.arpa/health returns
{"status":"OK"} when the service is live.
This page covers running archivebox-api as a long-lived service: the transports, a
Docker Compose stack, the optional agent server, putting it behind a Caddy reverse
proxy, and giving it a DNS name with Technitium. To provision the ArchiveBox
instance it connects to, see Backing Platform.
archivebox-apiships both an MCP server (console scriptarchivebox-mcp) and an A2A agent server (console scriptarchivebox-agent). The MCP server is the typed, deterministic tool surface; the agent server wraps it behind a Pydantic-AI graph agent and the Agent Web UI.
Run the MCP server¶
The transport is selected with --transport (or the TRANSPORT env var):
Health check (HTTP transports):
Configuration (environment)¶
archivebox-api is configured entirely from the environment. The required set:
| Var | Default | Meaning |
|---|---|---|
ARCHIVEBOX_BASE_URL |
http://localhost:8000 |
ArchiveBox API base URL |
ARCHIVEBOX_URL |
http://localhost:8000 |
Fallback alias for ARCHIVEBOX_BASE_URL |
ARCHIVEBOX_USERNAME |
(unset) | Username for authentication |
ARCHIVEBOX_PASSWORD |
(unset) | Password for authentication |
ARCHIVEBOX_API_KEY |
(unset) | API key for header authentication |
ARCHIVEBOX_TOKEN |
(unset) | Pre-configured authentication token |
ARCHIVEBOX_SSL_VERIFY |
False |
Verify TLS (self-signed homelab) |
AUTHENTICATIONTOOL |
True |
Register the authentication tool set |
CORETOOL |
True |
Register the core catalog tool set |
CLITOOL |
True |
Register the CLI tool set |
Plus HOST / PORT / TRANSPORT for HTTP transports, and the optional telemetry
(ENABLE_OTEL, OTEL_EXPORTER_OTLP_*) and access-governance (EUNOMIA_TYPE,
EUNOMIA_POLICY_FILE) settings. The full set is documented in
.env.example.
Copy it to .env and populate only what you use; the client remains inactive against
endpoints whose 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:
archivebox-api-mcp:
image: knucklessg1/archivebox-api:latest
container_name: archivebox-api-mcp
hostname: archivebox-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 ARCHIVEBOX_* values
docker compose -f docker/mcp.compose.yml up -d
docker compose -f docker/mcp.compose.yml logs -f
Agent server¶
For conversational and A2A operation, archivebox-api ships a Pydantic-AI graph
agent (console script archivebox-agent). It connects to the MCP server over
MCP_URL and exposes its own HTTP surface, including the Agent Web UI, on :9013.
export MCP_URL=http://localhost:8000/mcp
export PROVIDER=openai
export MODEL_ID=gpt-4o
archivebox-agent --host 0.0.0.0 --port 9013
The repo ships docker/agent.compose.yml,
which deploys the MCP server and the agent together and wires the agent to the MCP
server by container name:
archivebox-api-agent:
image: knucklessg1/archivebox-api:latest
container_name: archivebox-api-agent
hostname: archivebox-api-agent
restart: always
depends_on:
- archivebox-api-mcp
env_file:
- ../.env
command: [ "archivebox-agent" ]
environment:
- PYTHONUNBUFFERED=1
- HOST=0.0.0.0
- PORT=9013
- MCP_URL=http://archivebox-api-mcp:8000/mcp
- PROVIDER=${PROVIDER:-openai}
- MODEL_ID=${MODEL_ID:-gpt-4o}
- ENABLE_WEB_UI=True
ports:
- "9013:9013"
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
archivebox-api.arpa {
tls internal
reverse_proxy archivebox-api-mcp:8000
}
# Public — automatic Let's Encrypt
archivebox-api.example.com {
reverse_proxy archivebox-api-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=archivebox-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 archivebox-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": {
"archivebox-api": {
"command": "uv",
"args": ["run", "--package", "archivebox-api", "archivebox-mcp"],
"env": {
"ARCHIVEBOX_BASE_URL": "http://your-archivebox:8000",
"ARCHIVEBOX_USERNAME": "admin",
"ARCHIVEBOX_PASSWORD": "your-password"
}
}
}
}
For a remote HTTP server, point the client at http://archivebox-api.arpa/mcp instead.