mcparr
A self-hosted MCP gateway that lets AI assistants drive your media services like Radarr and Sonarr through a single authenticated endpoint.
README
mcparr
MCPify your *arr stack. mcparr is a self-hosted Model Context Protocol gateway that lets AI assistants (Claude, ChatGPT, GitHub Copilot, Cursor, ...) drive your media services - Radarr, Sonarr, and more - through a single, authenticated MCP endpoint.
What it does
mcparr runs as one Docker container exposing two ports:
- 7475 - config UI: a small web app to add service connections, test them, toggle tools, and copy the MCP connection details. Protected by an admin password.
- 7474 - MCP endpoint: a Streamable HTTP MCP server that AI clients connect to. Always requires a bearer token.
Each enabled service contributes a set of namespaced tools (for example
radarr_search_movies, sonarr_list_series). Toggling a service in the UI adds
or removes its tools live, without restarting the container.
Status
Early development. The v1 surface ships the gateway core plus Radarr and Sonarr; more services (Prowlarr, Plex, Overseerr/Seerr, Maintainerr, Lidarr) are planned.
Quick start
docker compose up -d
Then:
- Open
http://localhost:7475and set an admin password. - Add a service (base URL + API key), test the connection, and enable it.
- Open the Connect page to copy the MCP URL and bearer token into your AI client.
The config UI is published on localhost only by default. The MCP endpoint is exposed but always requires the token.
Connecting a client (VS Code / GitHub Copilot)
{
"servers": {
"mcparr": {
"type": "http",
"url": "http://your-host:7474/mcp",
"headers": { "Authorization": "Bearer <token-from-connect-page>" }
}
}
}
For Claude Desktop, use the mcp-remote bridge (snippet on the Connect page).
Hosted connectors (Claude.ai, ChatGPT) require HTTPS - put mcparr behind a TLS
reverse proxy.
Configuration
All state lives in the /data volume (SQLite database, encryption key, MCP
token, audit log). Useful environment variables:
| Variable | Default | Purpose |
|---|---|---|
MCPARR_HOST |
127.0.0.1 |
Bind address for both ports |
MCPARR_UI_PORT |
7475 |
Config UI port |
MCPARR_MCP_PORT |
7474 |
MCP endpoint port |
MCPARR_DATA_DIR |
/data |
State directory |
MCPARR_SECRET_KEY |
(generated) | Fernet key for secrets at rest |
MCPARR_MCP_TOKEN |
(generated) | MCP bearer token (env wins; rotation disabled) |
MCPARR_ADMIN_PASSWORD |
(unset) | Set the admin password headlessly on first run |
MCPARR_SEED_FILE |
<data>/seed.yaml |
Optional service seed file |
You can pre-populate services declaratively with a seed file - see config.yaml.example.
Security
The MCP endpoint always requires a token; the UI always requires a password. Service keys are encrypted at rest. See SECURITY.md and the threat model for details.
Development
python -m venv .venv && source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -e ".[dev]"
pytest -q
ruff check .
python -m mcparr.main # runs both servers locally
Translations are managed with Babel (babel.cfg); compile catalogs with
pybabel compile -d mcparr/ui/locales -D messages.
Adding a service
Create a ServiceModule subclass under mcparr/services/, declare its metadata
and tools, and decorate it with @register. The base class provides the HTTP
client, pagination, result shaping, and error mapping. Radarr and Sonarr are the
reference implementations.
License
See LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.