get_weekend_watchlistLABSCO SUMMARY
Film questions answered from TMDB's live catalogue rather than the model's memory.
The value is accuracy about things that change — what is trending this week, what rating a film currently carries, which id belongs to which of three films sharing a title — and the movie id that search returns is what makes the follow-up questions precise. Install it from the repository rather than through npx: the npm package is an older build that exposes only the search, trending and recommendation calls.
What it is
A bridge to The Movie Database API. It registers 24 tools — movie, TV and person search, trending, recommendations, where-to-watch lookups and watchlist planners — plus a resource URI for reading a film's full record as JSON.
What you get
- search_movies finds films by title or keywords, returning titles, TMDB ids, ratings and overviews
- get_trending returns what is trending, with the time window set to day or week
- get_recommendations takes a movie id and returns what TMDB recommends alongside it
- tmdb:///movie/<id> exposes a film's full record as a resource — cast, director, reviews and poster URL in JSON
- Beyond those three: watch-provider lookups, TV and person search, genre and keyword search, and workflow tools such as get_weekend_watchlist and build_franchise_watch_order
Requirements
A TMDB API key, free from your themoviedb.org account settings, placed in the .env file the repository provides. Installation is from source rather than a package manager: clone the repository, npm install, then npm run install:local writes the Codex and Claude Desktop configuration, after which the client needs a restart.
Setup effort
Build from source — git clone https://github.com/Laksh-star/mcp-server-tmdb.git
USE CASES
Finding the right film when several share a title
Search by title or keywords and read back the ids, years, ratings and overviews so the follow-up questions are about the right one.
Answering what is worth watching right now
Pull the trending list for the day or the week, rather than the model's impression of what is popular.
Finding something like a film you already liked
Take its id from a search and ask TMDB what it recommends next to it.
Getting the full record on one film
Read the resource for that id — cast, director, reviews and poster — as JSON instead of a summary paragraph.
24 tools
Listed in project code.
plan_watch_partybuild_franchise_watch_orderbuild_collection_gap_planrecommend_from_taste_profilebuild_release_calendar_watchlistbuild_person_watch_pathsearch_moviesget_trendingget_weekly_trending_by_languagesearch_by_genreadvanced_searchsearch_by_keywordget_movie_detailscompare_moviesget_recommendationsget_similar_moviesget_watch_providersfind_where_to_watchget_now_playingsearch_tv_showsget_trending_tvsearch_personget_person_detailsBEFORE YOU INSTALL
What to know first
Tools you get
24 tools registered in the project's code. Build it from the repository, as below. The npm package mcp-server-tmdb 1.0.0 is an older build with only search_movies, get_trending and get_recommendations.
Requires
- git and npm, for the clone-and-install steps
- TMDB_API_KEY in your shell environment or the repo .env file
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
Get a TMDB API key at https://www.themoviedb.org/ → Account Settings → API
Clone, install, and build
git clone https://github.com/Laksh-star/mcp-server-tmdb.git
cd mcp-server-tmdb
npm installCreate a local env file and add your TMDB key
cp .env.example .envInstall the local Codex and Claude Desktop integration
npm run install:localMCP Clients
Claude Desktop — local
{
"mcpServers": {
"tmdb-local": {
"command": "<YOUR_PATH_TO_MCP_SERVER_TMDB>/plugins/tmdb/scripts/run-server.sh",
"args": []
}
}
}Codex
[mcp_servers.tmdb_local]
command = "<YOUR_PATH_TO_MCP_SERVER_TMDB>/plugins/tmdb/scripts/run-server.sh"
[plugins."tmdb@openai-curated"]
enabled = true