Bazarr MCP server
Run and administer a Bazarr subtitle manager from Claude through 89 tools that mirror its entire REST API.
Subtitles: list available/embedded tracks, upload, download, delete, inspect contents, and run mods/tools (sync, translate, extract).
Wanted & blacklist: list wanted subtitles for movies and episodes, view and edit the subtitle blacklist.
Manual search & download: search providers for movie or episode subtitles and download a chosen result.
Providers: view provider status, reset providers, and filter provider history.
Library metadata: list movies, series, episodes, and their tags with paging and ID filters.
Language profiles: list languages, language profiles, and assign profiles to movies or series.
History & stats: browse movie/episode history events and history statistics by timeframe, action, provider, or language.
System administration: health, status, ping, releases, announcements (including dismiss), logs (list and rotate), and shutdown/restart.
Backups: list, create, restore, and delete backups.
Jobs & tasks: list the job queue, force-start/reorder/delete jobs, empty queues, list and run scheduled tasks.
Settings & auth: read or write the whole settings object and log in/out of the Bazarr UI.
Integrations: Radarr, Sonarr, Plex, and Jellyfin — browse their file systems, test connections, Plex OAuth/pin/webhook flows, and Jellyfin library and connection tests.
Inbound webhooks: trigger searches from Plex, Radarr, or Sonarr events and test external webhooks.
Large results: page, filter, and narrow oversized answers via
get_result_page.
Provides access to Bazarr's Jellyfin integration endpoints, enabling management of Jellyfin-related settings and operations through the Bazarr API.
Provides access to Bazarr's Plex integration endpoints, enabling management of Plex-related settings and operations through the Bazarr API.
Provides access to Bazarr's Radarr integration, allowing management of Radarr-linked movie libraries and subtitle workflows.
Provides access to Bazarr's Sonarr integration, allowing management of Sonarr-linked series libraries and subtitle workflows.
Click on "Deploy 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., "@Bazarr MCP server Find missing subtitles for my wanted movies and series."
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.
Bazarr MCP server
Run Bazarr from Claude.ai and Claude Code. All 89 operations of its API are tools. Not a curated subset: every endpoint Bazarr's web interface can reach, this can reach.
Using this server with a paid AI service costs money. Tool definitions and results are billed as input tokens, and an agent can call tools repeatedly on its own. You are responsible for every charge, so set spending limits with your provider. The author accepts no liability for any costs. SeeDISCLAIMER.md.
Why not the other options
Bazarr publishes no OpenAPI file, so the surface was read out of its source: 60 routed resources carrying 89 handlers.
Server | Bazarr tools | Coverage |
| 13 | 15 % |
| a handful | partial |
| subtitle verbs only | partial |
This one | 89 | 100 % |
The others cover wanted lists and a subtitle search. Nothing else exposes the blacklist, subtitle upload, the provider configuration, language profiles, announcements, backups, the task scheduler or the Plex and Jellyfin integrations.
Related MCP server: Arr Suite MCP Server
How it stays complete
Bazarr builds its Swagger at runtime from flask_restx decorators, so there is no file to download. scripts/extract_spec.py reads those decorators straight from the source and writes the equivalent OpenAPI document; scripts/generate_tools.py then turns it into tools:
git clone --depth 1 https://github.com/morpheus65535/bazarr.git /tmp/bazarr
python scripts/extract_spec.py /tmp/bazarr/bazarr/api openapi.json
python scripts/generate_tools.py openapi.json src/bazarr_mcp/tools.pyA test compares every generated call against every operation in the extracted spec, in both directions. An endpoint Bazarr adds and this misses fails the build; so does a tool pointing at an endpoint that does not exist.
Tool names
Verb first, derived from the method and path, so the name says what it does:
Pattern | Meaning | Example |
| Read a collection |
|
| Read one record |
|
| POST |
|
| PATCH |
|
| DELETE |
|
89 tools is a lot to put in front of a model at once. If your client supports tool filtering, narrow it to the groups you use.
What is covered
Every namespace: account, announcements, backups, badges, blacklist, endpoints, episodes, episodes_subtitles, files, files_radarr, files_sonarr, health, history, jobs, languages, languages_profiles, logs, movies, movies_subtitles, notifications, oauth, ping, plex, providers, providers_episodes, providers_movies, radarr, releases, searches, series, settings, sonarr, stats, status, subtitles, subtitles_contents, subtitles_info, system, tags, tasks, wanted.
Setup
git clone https://github.com/rollecode/bazarr-mcp.git
cd bazarr-mcp
uv venv && uv pip install -e .export BAZARR_URL=http://127.0.0.1:6767
export BAZARR_API_KEY=... # Settings, General, Security, API keyClaude Code
claude mcp add bazarr -- /path/to/bazarr-mcp/.venv/bin/bazarr-mcpWriting
Bazarr takes form fields rather than JSON bodies, and these tools send them that way. Settings are read and written whole through list_system_settings and create_system_settings.
Hosting it
Running it over HTTP puts it in reach of Claude.ai as a custom connector, and of Claude Code on other machines. Three tiers, the same shape the other servers in this family use:
Tier | Port | What it does |
| 8560 | The server. No login of its own, never exposed |
nginx | 8561 | Front door, behind a Cloudflare Tunnel |
| 8562 | OAuth 2.1 sign-in, or a fixed bearer token |
npm install
node set-password.js 'a password for the sign-in page'
printf 'BAZARR_URL=...\n' > ~/.config/bazarr-mcp/env
chmod 600 ~/.config/bazarr-mcp/envCopy systemd/*.service into /etc/systemd/system/, replacing YOUR_USER and the ISSUER hostname, then:
sudo systemctl enable --now bazarr-mcp bazarr-mcp-authPoint nginx/bazarr-mcp.conf at your own hostname and send the tunnel at 127.0.0.1:8561.
Environment the server itself reads: BAZARR_URL, BAZARR_API_KEY. The sign-in page carries the Bazarr mark and accent colour, set through APP_NAME, APP_ACCENT and APP_BLURB in the auth unit.
Claude.ai
Settings, Connectors, Add custom connector, URL https://bazarr-mcp.your-domain/mcp, client ID and secret blank. The sign-in page asks for the password set above. Connectors belong to the account, so adding it once covers mobile too.
Development
uv pip install -e . pytest ruff
.venv/bin/python -m pytest tests
.venv/bin/ruff check .Available Tools
90 toolscreate_episodes_blacklistCIdempotent
Add an episodes subtitles to blacklist.
POST /api/episodes/blacklist
Args: seriesid: Series ID episodeid: Episode ID provider: Provider name subs_id: Subtitles ID language: Subtitles language subtitles_path: Subtitles file path
Inputs: subs_id (required), language (required), provider (required), seriesid (required), +2 more. Returns: result
create_episodes_subtitlesBIdempotent
Upload an episode subtitles.
POST /api/episodes/subtitles
Args: seriesid: Series ID episodeid: Episode ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string file: Subtitles file as file upload object
Inputs: hi (required), file (required), forced (required), language (required), +2 more. Returns: result
create_jellyfin_test_connectionAIdempotent
Test connection to a Jellyfin server with provided credentials.
POST /api/jellyfin/test-connection
Args: url: Jellyfin server URL apikey: Jellyfin API key
Inputs: url (required), apikey (required). Returns: result
create_moviesBIdempotent
Update specific movies languages profile.
POST /api/movies
Args: radarrid: Radarr movie(s) ID profileid: Languages profile(s) ID or "none"
Inputs: radarrid, profileid. Returns: result
create_movies_blacklistCIdempotent
Add a movies subtitles to blacklist.
POST /api/movies/blacklist
Args: radarrid: Radarr ID provider: Provider name subs_id: Subtitles ID language: Subtitles language subtitles_path: Subtitles file path
Inputs: subs_id (required), language (required), provider (required), radarrid (required), +1 more. Returns: result
create_movies_subtitlesBIdempotent
Upload a movie subtitles.
POST /api/movies/subtitles
Args: radarrid: Movie ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string file: Subtitles file as file upload object
Inputs: hi (required), file (required), forced (required), language (required), +1 more. Returns: result
create_plex_apikeyCIdempotent
POST plex/apikey.
POST /api/plex/apikey
Args: apikey: API key
Inputs: apikey (required). Returns: result
create_plex_encrypt_apikeyCIdempotent
POST plex/encrypt-apikey.
POST /api/plex/encrypt-apikey
Inputs: no named parameters. Returns: result
create_plex_oauth_logoutCIdempotent
POST plex/oauth/logout.
POST /api/plex/oauth/logout
Inputs: no named parameters. Returns: result
create_plex_oauth_pinCIdempotent
POST plex/oauth/pin.
POST /api/plex/oauth/pin
Args: client_id: Client ID
Inputs: client_id. Returns: result
create_plex_select_serverDIdempotent
POST plex/select-server.
POST /api/plex/select-server
Args: machine_identifier: Machine identifier name: Server name uri: Connection URI local: Is local connection connections: All available connection URIs
Inputs: uri (required), name (required), machine_identifier (required), local, +1 more. Returns: result
create_plex_test_connectionCIdempotent
POST plex/test-connection.
POST /api/plex/test-connection
Args: uri: Server URI
Inputs: uri (required). Returns: result
create_plex_webhook_createDIdempotent
POST plex/webhook/create.
POST /api/plex/webhook/create
Inputs: no named parameters. Returns: result
create_plex_webhook_deleteCIdempotent
POST plex/webhook/delete.
POST /api/plex/webhook/delete
Args: webhook_url: Webhook URL to delete
Inputs: webhook_url (required). Returns: result
create_providersBIdempotent
Reset providers status.
POST /api/providers
Args: action: Action to perform from ["reset"]
Inputs: action (required). Returns: result
create_providers_episodesBIdempotent
Manually download an episode subtitles.
POST /api/providers/episodes
Args: seriesid: Series ID episodeid: Episode ID hi: HI subtitles from ["True", "False"] forced: Forced subtitles from ["True", "False"] original_format: Use original subtitles format from ["True", "False"] provider: Provider name subtitle: Subtitle ID as returned by GET
Inputs: hi (required), forced (required), provider (required), seriesid (required), +3 more. Returns: result
create_providers_moviesAIdempotent
Manually download a movie subtitles.
POST /api/providers/movies
Args: radarrid: Movie ID hi: HI subtitles from ["True", "False"] forced: Forced subtitles from ["True", "False"] original_format: Use original subtitles format from ["True", "False"] provider: Provider name subtitle: Subtitle ID as returned by GET
Inputs: hi (required), forced (required), provider (required), radarrid (required), +2 more. Returns: result
create_seriesBIdempotent
Update specific series languages profile.
POST /api/series
Args: seriesid: Sonarr series ID profileid: Languages profile(s) ID or "none"
Inputs: seriesid, profileid. Returns: result
create_systemAIdempotent
Shutdown or restart Bazarr.
POST /api/system
Args: action: Action to perform from ["shutdown", "restart"]
Inputs: action (required). Returns: result
create_system_accountAIdempotent
Login or logout from Bazarr UI when using form login.
POST /api/system/account
Args: action: Action from ["login", "logout"] username: Bazarr username password: Bazarr password
Inputs: action (required), password, username. Returns: result
create_system_announcementsAIdempotent
Mark announcement as dismissed.
POST /api/system/announcements
Args: hash: hash of the announcement to dismiss
Inputs: hash (required). Returns: result
create_system_backupsBIdempotent
Create a new backup.
POST /api/system/backups
Inputs: no named parameters. Returns: result
create_system_jobsAIdempotent
Force start, move to top or move to bottom of the queue a specific job.
POST /api/system/jobs
Args: id: Job ID act onto action: Action to perform from ["force_start", "move_top", "move_bottom"]
Inputs: id (required), action (required). Returns: result
create_system_settingsCIdempotent
POST system/settings.
POST /api/system/settings
Inputs: no named parameters. Returns: result
create_system_tasksCIdempotent
Run task.
POST /api/system/tasks
Args: taskid: Task id of the task to run
Inputs: taskid (required). Returns: result
create_system_webhooks_testAIdempotent
Test external webhook connection.
POST /api/system/webhooks/test
Inputs: no named parameters. Returns: result
create_webhooks_plexBIdempotent
Trigger subtitles search on play media event in Plex.
POST /api/webhooks/plex
Args: payload: Webhook payload
Inputs: payload (required). Returns: result
create_webhooks_radarrAIdempotent
Search for missing subtitles based on Radarr webhooks.
POST /api/webhooks/radarr
Inputs: no named parameters. Returns: result
create_webhooks_sonarrAIdempotent
Search for missing subtitles based on Sonarr webhooks.
POST /api/webhooks/sonarr
Inputs: no named parameters. Returns: result
delete_episodes_blacklistCDestructiveIdempotent
Delete an episodes subtitles from blacklist.
DELETE /api/episodes/blacklist
Args: all: Empty episodes subtitles blacklist provider: Provider name subs_id: Subtitles ID
Inputs: all, subs_id, provider. Returns: result
delete_episodes_subtitlesBDestructiveIdempotent
Delete an episode subtitles.
DELETE /api/episodes/subtitles
Args: seriesid: Series ID episodeid: Episode ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string path: Path of the subtitles file
Inputs: hi (required), path (required), forced (required), language (required), +2 more. Returns: result
delete_movies_blacklistBDestructiveIdempotent
Delete a movies subtitles from blacklist.
DELETE /api/movies/blacklist
Args: all: Empty movies subtitles blacklist provider: Provider name subs_id: Subtitles ID
Inputs: all, subs_id, provider. Returns: result
delete_movies_subtitlesCDestructiveIdempotent
Delete a movie subtitles.
DELETE /api/movies/subtitles
Args: radarrid: Movie ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string path: Path of the subtitles file
Inputs: hi (required), path (required), forced (required), language (required), +1 more. Returns: result
delete_system_backupsADestructiveIdempotent
Delete a backup file.
DELETE /api/system/backups
Args: filename: Backups to delete filename
Inputs: filename (required). Returns: result
delete_system_jobsADestructiveIdempotent
Delete a job from the queue.
DELETE /api/system/jobs
Args: id: Job ID to delete from queue
Inputs: id (required). Returns: result
delete_system_logsADestructiveIdempotent
Force log rotation and create a new log file.
DELETE /api/system/logs
Inputs: no named parameters. Returns: result
get_plex_oauth_pin_by_pin_id_checkCRead-onlyIdempotent
GET plex/oauth/pin/string:pin_id/check.
GET /api/plex/oauth/pin/{pin_id}/check
Args: pin_id: Path parameter.
Inputs: pin_id (required). Returns: result
get_result_pageARead-onlyIdempotent
Read more of a result that was too large to return in one piece.
Any tool whose answer is too large returns its first page with a
paging block. Pass that block's result_id here. Nothing is fetched
again: the stored result is paged, filtered and narrowed.
Args:
result_id: From the paging block of the earlier answer.
offset: First item to return, usually the previous next_offset.
limit: At most this many items. Fewer come back if they would not fit.
fields: Only these keys of each item. Dots reach nested keys, such as
"statistics.sizeOnDisk". The paging.fields list shows what exists.
match: Keep only items whose field contains this text, ignoring case,
for example {"title": "alien", "year": "1979"}.
path: Which list to page when the result holds several, as named in
paging.lists. Defaults to the largest.
Inputs: result_id (required), path, limit, match, +2 more. Returns: result
list_badgesARead-onlyIdempotent
Get badges count to update the UI.
GET /api/badges
Inputs: no named parameters. Returns: result
list_episodesBRead-onlyIdempotent
List episodes metadata for specific series or episodes.
GET /api/episodes
Args: seriesid: Series IDs to list episodes for episodeid: Episodes ID to list
Inputs: seriesid, episodeid. Returns: result
list_episodes_blacklistARead-onlyIdempotent
List blacklisted episodes subtitles.
GET /api/episodes/blacklist
Args: start: Paging start integer length: Paging length integer
Inputs: start, length. Returns: result
list_episodes_historyBRead-onlyIdempotent
List episodes history events.
GET /api/episodes/history
Args: start: Paging start integer length: Paging length integer episodeid: Episode ID
Inputs: start, length, episodeid. Returns: result
list_episodes_wantedBRead-onlyIdempotent
List episodes wanted subtitles.
GET /api/episodes/wanted
Args: start: Paging start integer length: Paging length integer episodeid: Episodes ID to list
Inputs: start, length, episodeid. Returns: result
list_filesBRead-onlyIdempotent
List Bazarr file system content.
GET /api/files
Args: path: Path to browse
Inputs: path. Returns: result
list_files_radarrBRead-onlyIdempotent
List Radarr file system content.
GET /api/files/radarr
Args: path: Path to browse
Inputs: path. Returns: result
list_files_sonarrARead-onlyIdempotent
List Sonarr file system content.
GET /api/files/sonarr
Args: path: Path to browse
Inputs: path. Returns: result
list_history_statsARead-onlyIdempotent
Get history statistics.
GET /api/history/stats
Args: time_frame: Timeframe to get stats for. Must be in ["week", "month", "trimester", "year"] action: Action type to filter for. provider: Provider name to filter for. language: Language name to filter for
Inputs: action, language, provider, time_frame. Returns: result
list_jellyfin_librariesARead-onlyIdempotent
List available movie and series libraries from the Jellyfin server.
GET /api/jellyfin/libraries
Args: url: Jellyfin server URL apikey: Jellyfin API key
Inputs: url, apikey. Returns: result
list_moviesBRead-onlyIdempotent
List movies metadata for specific movies.
GET /api/movies
Args: start: Paging start integer length: Paging length integer radarrid: Movies IDs to get metadata for
Inputs: start, length, radarrid. Returns: result
list_movies_blacklistBRead-onlyIdempotent
List blacklisted movies subtitles.
GET /api/movies/blacklist
Args: start: Paging start integer length: Paging length integer
Inputs: start, length. Returns: result
list_movies_historyBRead-onlyIdempotent
List movies history events.
GET /api/movies/history
Args: start: Paging start integer length: Paging length integer radarrid: Movie ID
Inputs: start, length, radarrid. Returns: result
list_movies_tagsARead-onlyIdempotent
List all distinct tags assigned to any movie.
GET /api/movies/tags
Inputs: no named parameters. Returns: result
list_movies_wantedCRead-onlyIdempotent
List movies wanted subtitles.
GET /api/movies/wanted
Args: start: Paging start integer length: Paging length integer radarrid: Movies ID to list
Inputs: start, length, radarrid. Returns: result
list_plex_autopulse_configBRead-onlyIdempotent
GET plex/autopulse/config.
GET /api/plex/autopulse/config
Inputs: no named parameters. Returns: result
list_plex_oauth_librariesCRead-onlyIdempotent
GET plex/oauth/libraries.
GET /api/plex/oauth/libraries
Inputs: no named parameters. Returns: result
list_plex_oauth_pinDRead-onlyIdempotent
GET plex/oauth/pin.
GET /api/plex/oauth/pin
Inputs: no named parameters. Returns: result
list_plex_oauth_serversBRead-onlyIdempotent
GET plex/oauth/servers.
GET /api/plex/oauth/servers
Inputs: no named parameters. Returns: result
list_plex_oauth_validateCRead-onlyIdempotent
GET plex/oauth/validate.
GET /api/plex/oauth/validate
Inputs: no named parameters. Returns: result
list_plex_select_serverCRead-onlyIdempotent
GET plex/select-server.
GET /api/plex/select-server
Inputs: no named parameters. Returns: result
list_plex_test_connectionCRead-onlyIdempotent
GET plex/test-connection.
GET /api/plex/test-connection
Inputs: no named parameters. Returns: result
list_plex_webhook_listCRead-onlyIdempotent
GET plex/webhook/list.
GET /api/plex/webhook/list
Inputs: no named parameters. Returns: result
list_providersBRead-onlyIdempotent
Get providers status.
GET /api/providers
Args: history: Provider name for history stats
Inputs: history. Returns: result
list_providers_episodesBRead-onlyIdempotent
Search manually for an episode subtitles.
GET /api/providers/episodes
Args: episodeid: Episode ID
Inputs: episodeid. Returns: result
list_providers_moviesCRead-onlyIdempotent
Search manually for a movie subtitles.
GET /api/providers/movies
Args: radarrid: Movie ID
Inputs: radarrid. Returns: result
list_seriesCRead-onlyIdempotent
List series metadata for specific series.
GET /api/series
Args: start: Paging start integer length: Paging length integer seriesid: Series IDs to get metadata for
Inputs: start, length, seriesid. Returns: result
list_series_tagsARead-onlyIdempotent
List all distinct tags assigned to any series.
GET /api/series/tags
Inputs: no named parameters. Returns: result
list_subtitlesBRead-onlyIdempotent
Return available audio and embedded subtitles tracks with external subtitles. Used for manual subsync.
GET /api/subtitles
Args: subtitles_path: External subtitles file path sonarr_episode_id: Sonarr Episode ID radarr_movie_id: Radarr Movie ID
Inputs: subtitles_path, radarr_movie_id, sonarr_episode_id. Returns: result
list_subtitles_contentsCRead-onlyIdempotent
Retrieve subtitle file contents.
GET /api/subtitles/contents
Args: subtitle_path: Subtitle filepath
Inputs: subtitle_path. Returns: result
list_subtitles_infoBRead-onlyIdempotent
Guessit over subtitles filename.
GET /api/subtitles/info
Args: filenames: Subtitles filenames
Inputs: filenames. Returns: result
list_system_announcementsARead-onlyIdempotent
List announcements relative to Bazarr.
GET /api/system/announcements
Inputs: no named parameters. Returns: result
list_system_backupsARead-onlyIdempotent
List backup files.
GET /api/system/backups
Inputs: no named parameters. Returns: result
list_system_healthBRead-onlyIdempotent
List health issues.
GET /api/system/health
Inputs: no named parameters. Returns: result
list_system_jobsBRead-onlyIdempotent
List jobs from the queue.
GET /api/system/jobs
Args: id: Job ID to return status: Job status to return
Inputs: id, status. Returns: result
list_system_languagesARead-onlyIdempotent
List languages for history filter or for language filter menu.
GET /api/system/languages
Args: history: Language name for history stats
Inputs: history. Returns: result
list_system_languages_profilesARead-onlyIdempotent
List languages profiles.
GET /api/system/languages/profiles
Inputs: no named parameters. Returns: result
list_system_logsARead-onlyIdempotent
List log entries.
GET /api/system/logs
Inputs: no named parameters. Returns: result
list_system_pingBRead-onlyIdempotent
Return status and http 200.
GET /api/system/ping
Inputs: no named parameters. Returns: result
list_system_releasesBRead-onlyIdempotent
Get Bazarr releases.
GET /api/system/releases
Inputs: no named parameters. Returns: result
list_system_searchesARead-onlyIdempotent
List results from query.
GET /api/system/searches
Args: query: Series or movie name to search for
Inputs: query. Returns: result
list_system_settingsBRead-onlyIdempotent
GET system/settings.
GET /api/system/settings
Inputs: no named parameters. Returns: result
list_system_statusARead-onlyIdempotent
Return environment information and versions.
GET /api/system/status
Inputs: no named parameters. Returns: result
list_system_tasksBRead-onlyIdempotent
List tasks.
GET /api/system/tasks
Args: taskid: List tasks or a single task properties
Inputs: taskid. Returns: result
patch_episodes_subtitlesCIdempotent
Download an episode subtitles.
PATCH /api/episodes/subtitles
Args: seriesid: Series ID episodeid: Episode ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string
Inputs: hi (required), forced (required), language (required), seriesid (required), +1 more. Returns: result
patch_moviesBIdempotent
Run actions on specific movies.
PATCH /api/movies
Args: radarrid: Radarr movie ID action: Action to perform from ["scan-disk", "search-missing", "search-wanted", "sync"]
Inputs: action, radarrid. Returns: result
patch_movies_subtitlesBIdempotent
Download a movie subtitles.
PATCH /api/movies/subtitles
Args: radarrid: Movie ID language: Language code2 forced: Forced true/false as string hi: HI true/false as string
Inputs: hi (required), forced (required), language (required), radarrid (required). Returns: result
patch_seriesAIdempotent
Run actions on specific series.
PATCH /api/series
Args: seriesid: Sonarr series ID action: Action to perform from ["scan-disk", "search-missing", "search-wanted", "sync"]
Inputs: action, seriesid. Returns: result
patch_subtitlesAIdempotent
Apply mods/tools on external subtitles.
PATCH /api/subtitles
Args: action: Action from ["sync", "translate", "extract" or mods name] language: Language code2 path: Subtitles file path type: Media type from ["episode", "movie"] id: Media ID (episodeId, radarrId) forced: Forced subtitles from ["True", "False"] hi: HI subtitles from ["True", "False"] original_format: Use original subtitles format from ["True", "False"] reference: Reference to use for sync from video file track number (a:0) or some subtitles file path max_offset_seconds: Maximum offset seconds to allow no_fix_framerate: Don't try to fix framerate from ["True", "False"] gss: Use Golden-Section Search from ["True", "False"] subtitles_id: Subtitles database ID (required for the "extract" action)
Inputs: id (required), type (required), action (required), language (required), +9 more. Returns: result
patch_system_backupsCIdempotent
Restore a backup file.
PATCH /api/system/backups
Args: filename: Backups to restore filename
Inputs: filename (required). Returns: result
patch_system_jobsBIdempotent
Empty a specific jobs queue.
PATCH /api/system/jobs
Args: queue_name: Jobs queue name to empty
Inputs: queue_name (required). Returns: result
patch_system_notificationsBIdempotent
Test a notifications provider URL.
PATCH /api/system/notifications
Args: url: Notifications provider URL
Inputs: url (required). Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 90 tools
Many tools share similar resource names (e.g., list_movies, list_movies_blacklist, list_movies_history, list_movies_wanted; create_movies vs patch_movies) and some names are misleading (create_movies updates a languages profile, not creates a movie). An agent must read endpoint paths carefully to avoid misselection.
Names follow a consistent HTTP-method-prefixed snake_case pattern (list_, create_, delete_, patch_) derived from endpoint paths. Minor deviations exist, such as get_plex_oauth_pin_by_pin_id_check using get_ instead of list_, and redundant names like create_plex_webhook_create.
90 tools is far beyond the 3-15 sweet spot and exceeds the 50+ threshold for extreme mismatch. While it mirrors the full Bazarr API, the count is unwieldy for an agent to navigate effectively.
The surface covers system, series, episodes, movies, subtitles, providers, blacklist, history, wanted, files, and integrations like Plex and Jellyfin, with CRUD-like operations for most entities. Minor gaps exist (e.g., no delete_series or delete_movies for media items), but core workflows are supported.
Maintenance
Related MCP Connectors
Generate SRT and VTT subtitles for any video or audio file, 99 languages.
Subtitle engine for AI agents: one upload gives a written track and a translation, timings exact.
Transcribe audio and video, and translate subtitles into 50+ languages.
Convert, clean, retime and validate SRT, WebVTT, ASS/SSA, SBV and TTML subtitles. 8 tools.
Related MCP Servers
- AlicenseBqualityDmaintenanceEnables searching and downloading subtitles from OpenSubtitles.com through their API. Supports comprehensive search by movie title, IMDB/TMDB ID, file hash, and TV show episodes with multiple subtitle formats (SRT, ASS, VTT).32MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to manage media automation services like Sonarr, Radarr, Prowlarr, Bazarr, Overseerr, and Plex through natural language commands.9MIT
- AlicenseCqualityBmaintenanceEnables full control of Sonarr from Claude.ai and Claude Code by exposing all 234 v3 API operations as tools for managing media libraries.23523 npmMIT
- AlicenseCqualityBmaintenanceEnables managing all Radarr movie management operations via natural language, covering 237 v3 API endpoints for tasks like managing movies, queues, quality profiles, and custom formats with full API completeness.23863 npmMIT