media-mcp
Manages audiobooks and ebooks, including searching releases via Prowlarr, checking download status and import, grabbing releases, and canceling downloads.
Enables downloading Bandcamp albums or discographies directly via yt-dlp, separate from torrent-based music downloads.
Provides tools to scan libraries, refresh items, check media availability, and view recently added content in Jellyfin.
Provides management of qBittorrent torrents, allowing listing, pausing, resuming, and deleting torrents.
Integrates with Radarr for movie management, including viewing queue, searching releases, grabbing releases, and deleting movies.
Integrates with Sonarr for TV series management, including queue, release search, grab, and delete operations.
Sends push notifications for webhook events (e.g., import completion) to a Telegram chat.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@media-mcpsearch for the movie Interstellar"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
media-mcp
MCP server for media stack management — movies/TV (Jellyseerr, Radarr, Sonarr, Jellyfin), audiobooks/ebooks (Prowlarr → qBittorrent → Audiobookshelf), and music (Prowlarr/Bandcamp → qBittorrent → Navidrome). Shared by Sebastian, Max, and Claude Code.
Where it runs
Host: the bot host
Path:
~/projects/media-mcp/Transport: stdio via
uvx(MCP) + HTTP :8097 (webhook receiver)Deploy: CI on push (git pull + restart of the webhook service)
Media stack: a media host running Jellyseerr/Radarr/Sonarr/Prowlarr/qBittorrent/Jellyfin
Related MCP server: Mediabox MCP
Architecture
media host bot host dev machine
┌──────────────┐ ┌────────────────┐ ┌──────────────┐
│ Jellyseerr │◄──HTTP──┐ │ server.py │ │ media-mcp │
│ Radarr │◄────────┤ │ (FastMCP) │ │ (full mode) │
│ Sonarr │◄────────┤ │ MODE=safe|full│ └──────────────┘
│ Prowlarr │◄────────┤ └──┬──────────┬──┘
│ qBittorrent │◄────────┤ │ │
│ Jellyfin │◄────────┘ ┌──▼──┐ ┌───▼────┐
└──────────────┘ │ Max │ │Sebast. │
│ │safe │ │ full │
│ webhook └─────┘ └────────┘
└─────────────────►webhook_receiver.py :8097
→ Telegram pushModes
Mode | Consumer | Tools |
| Max (secondary user bot) | movie/TV search+request+status, |
| Sebastian, Claude Code | all safe + |
Set via MEDIA_MCP_MODE env variable.
Tools
SAFE (11 tools)
Movies / TV:
Tool | Description |
| Search movies/TV by name (Jellyseerr/TMDb) |
| Check if available in Jellyfin |
| Order a movie/show (HD-1080p, Russian audio) |
| Show request statuses |
| What's new in Jellyfin |
| Download progress (Radarr/Sonarr queue) |
| Recommendations based on a title |
Audiobooks / ebooks (Prowlarr → qBittorrent → Audiobookshelf):
Tool | Description |
| Search audiobook/ebook releases via Prowlarr |
| Download status + whether imported into Audiobookshelf |
| Recently added books in Audiobookshelf |
| Grab a release, route to |
FULL (additional 23 tools)
Radarr: radarr_queue, radarr_search_releases, radarr_grab_release, radarr_delete_movie
Sonarr: sonarr_queue, sonarr_search_releases, sonarr_grab_release, sonarr_delete_series
qBit: qbit_list_torrents, qbit_pause, qbit_resume, qbit_delete
Prowlarr: prowlarr_indexer_status, prowlarr_test_indexer
Jellyfin: jellyfin_scan_library, jellyfin_refresh_item
Books: book_cancel
Music (Prowlarr/Bandcamp → qBittorrent → Navidrome):
Tool | Description |
| Search music releases via Prowlarr |
| Download status + whether imported into Navidrome |
| Recently added albums in Navidrome |
| Grab a release, route to |
| Cancel a music download |
| Download a Bandcamp album/discography via yt-dlp (separate from torrents) |
See discovery/ for the batch taste-based music-collection top-up that drives music_*.
Destructive tools (*_delete, *_grab, *_cancel) require confirm=True. Default is dry-run preview.
Run
# SAFE mode
MEDIA_MCP_MODE=safe JELLYSEERR_URL=http://localhost:5055 JELLYSEERR_API_KEY=... \
uvx --with mcp --with requests python server.py
# FULL mode
MEDIA_MCP_MODE=full JELLYSEERR_URL=... RADARR_URL=... RADARR_API_KEY=... \
uvx --with mcp --with requests python server.pyEnvironment Variables
Variable | Required for | Description |
| all |
|
| all | Jellyseerr URL |
| all | Jellyseerr API key |
| safe+full | Radarr URL (used for progress in safe) |
| safe+full | Radarr API key |
| safe+full | Sonarr URL (used for progress in safe) |
| safe+full | Sonarr API key |
| safe+full | Jellyfin URL |
| safe+full | Jellyfin API key |
| full | Prowlarr URL |
| full | Prowlarr API key |
| full | qBittorrent URL |
| full | qBittorrent username |
| full | qBittorrent password |
Books add ABS_URL / ABS_API_KEY (Audiobookshelf); music adds NAVIDROME_URL /
NAVIDROME_USER / NAVIDROME_PASS. See .env.example for the full
list (importer paths, Telegram notifications, webhook, discovery tuning).
Stack
Python 3.11+
FastMCP (
mcp.server.fastmcp)requests
No Docker — runs directly via uvx
Status
Movies/TV, audiobooks/ebooks, and music tools implemented (34 total). Webhook
receiver live. Importers (importer.py, music_importer.py) run on the media host.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/MikeSviblov/media-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server