arr-mcp
A unified MCP server for your self-hosted media stack (Radarr, Sonarr, Jellyfin, etc.). It correlates across services to answer questions no single service can, with safety features like untrusted-output fencing and safe-by-default deletions.
README
arr-mcp
One MCP server for your whole self-hosted media stack — not one per service.
Ask questions no single service can answer. "Why isn't the film I requested on Tuesday showing up in Jellyfin?" spans Seerr, Radarr, Prowlarr, SABnzbd and Jellyfin. arr-mcp correlates them and gives you the causal chain.
- A web config page that diagnoses connections instead of printing pass/fail, and shows live logs while you debug.
- Tool output is treated as untrusted data, never instruction. Release names from public indexers are attacker-controllable and flow straight into model context; arr-mcp fences them.
- Safe by default. Deletion is off until you deliberately enable it, and then still asks per call.
Status: 0.1 walking skeleton
Radarr and a single
stack_healthtool. Not yet useful — this is the foundation the rest lands on. See the roadmap.
Planned services
Radarr · Sonarr · Prowlarr · Bazarr · Jellyfin · Seerr · SABnzbd · Transmission
Quick start
Not yet — there is no published image until 0.1.0 ships. Once it does:
services:
arr-mcp:
image: ghcr.io/bardesss/arr-mcp:latest
ports: ['6060:6060']
volumes: ['./config:/config']
environment:
- PUID=1000
- PGID=1000
- TZ=Europe/Amsterdam
restart: unless-stopped
The bearer token for the MCP endpoint is generated on first run and printed to
the container log. Until the config UI lands in 0.5, edit
config/config.yaml by hand and restart the container to pick up changes.
Roadmap
| Version | Delivers |
|---|---|
| 0.1 | Walking skeleton: stateless MCP transport, bearer auth, Radarr, stack_health |
| 0.2 | The remaining seven service adapters and 11 read tools |
| 0.3 | Cross-service correlation: identity resolver, three-way library join, diagnose |
| 0.4 | Writes: permission tiers, dry_run, write audit, per-call confirmation |
| 0.5 | Web config page: dashboard, diagnosing connection tests, log streams |
| 0.6 → 1.0 | Metadata providers, MCP resources and prompts |
Each version is a self-contained, shippable slice — the goal is that 0.3 already answers questions no individual service can, with 0.4–0.6 making it complete.
Requirements
- A LAN-reachable install of at least one supported service
- Docker, or Node 24+ to run from source
- An MCP client speaking protocol revision
2026-07-28
Security
arr-mcp is not designed to be exposed to the internet. The /mcp endpoint
requires a bearer token because "LAN-only" is a network assumption rather than a
security control — the endpoint fronts up to eight API keys and, once enabled,
file deletion, and a home network contains guest phones and IoT devices.
Thanks
arr-mcp is glue. The hard parts belong to other people.
The services it speaks to — none of this exists without them, and every one is free software maintained largely by volunteers:
Radarr · Sonarr · Prowlarr · Bazarr · Jellyfin · Seerr · SABnzbd · Transmission
The libraries it is built on:
| MCP TypeScript SDK | the protocol, and a stateless transport that made this simple |
| Hono | the HTTP server |
| Zod | config validation and tool input schemas |
| Pino | logging |
| Vitest | tests |
| better-sqlite3 | the log ring buffer and write audit |
| yaml | reading config.yaml |
| TypeScript | the language |
If you find arr-mcp useful, consider supporting the services above first.
Licence
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.