torrent-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JACKETT_URL | Yes | Base URL of your Jackett instance. | |
| OMDB_API_KEY | Yes | Your OMDb API key used for title lookups. | |
| JACKETT_API_KEY | Yes | Your Jackett API key. | |
| TORRENT_MCP_PROFILES | Yes | Configuration for one or more qBittorrent+Jackett RSS queue profiles (movies, series, upgrades, etc.). | |
| DOCKER_RESTART_ENABLED | No | Set to 'true' to allow the server to restart Docker containers by name from your own configured profiles. | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_profilesA | List the configured queue profile names (e.g. "movies", "series"). |
| find_imdb_candidatesB | Scrape an IMDb list page (e.g. a "coming this year" list) for title ids. |
| check_candidatesB | Split a list of IMDb ids into already-queued, already-in-Plex, and new.
|
| add_to_queueA | Add an IMDb id to a queue profile's feeds.json so qBittorrent starts searching for it.
|
| remove_from_queueB | Remove an IMDb id from a queue profile's feeds.json. |
| list_queueA | List a queue profile's entries. Set resolve_titles=True to include title/year via OMDb (slower). |
| switch_indexerA | Repoint every feed URL in a queue profile from one Jackett indexer to another. Set |
| list_indexersB | List Jackett indexers known to the server. |
| check_imdb_search_supportB | Check whether a Jackett indexer supports searching by IMDb id. |
| search_torrentsA | Search one Jackett indexer (or "all") for torrents matching an IMDb id. |
| find_largest_torrentA | Find the largest-sized search result for an IMDb id on one indexer (highest quality heuristic). |
| lookup_titleA | Resolve an IMDb id to its title/year/type via OMDb. |
| search_titleB | Fuzzy-search OMDb by title, optionally narrowed by year. |
| identify_titleB | Guess the IMDb id/title for a messy title (e.g. cleaned-up filename), via OMDb. |
| find_duplicate_filenamesB | Find likely-duplicate filenames in a media directory (e.g. movie.avi vs movie.mp4). |
| diff_missing_urlsA | List IMDb title URLs present in an old library export but missing from a newer one. |
| archive_downloadsA | Move finished downloads (by extension) out of the download client's folders into the archive. Falls back to ARCHIVE_SOURCE_DIRS/ARCHIVE_DEST_DIR from the server's environment when not given explicitly. |
| backup_queueB | Back up a queue profile's feeds.json to a timestamped file. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 18 tools
Most tools are clearly separated by object (profiles, queue entries, indexers, titles, files), and descriptions specify exact inputs and outputs. The main potential confusion is between lookup_title/search_title/identify_title and search_torrents/find_largest_torrent, but their boundary conditions are stated clearly enough.
Every tool uses a consistent snake_case, imperative verb-first naming pattern such as list, add, remove, search, find, or check. Even longer names like check_imdb_search_support and find_duplicate_filenames follow the same convention, making the set predictable.
At 18 tools, this is slightly above the typical sweet spot, but the server spans queue management, indexer operations, metadata resolution, and media-file housekeeping. Each tool maps to a distinct operation, so the count feels broad rather than redundant.
The core pipeline—finding IMDb candidates, checking them against the queue/Plex, adding and removing queue entries, searching indexers, and archiving downloads—is well covered. Minor gaps such as profile creation/deletion and directly adding a found torrent to qBittorrent exist but can be worked around with the available tools.