MCP TMDB
An implementation of the Model Context Protocol that provides tools to interact with The Movie Database (TMDB), allowing users to search for movie collaborations, actor filmographies, and other film industry data.
README
MCP TMDB
This project is an implementation of the Model Context Protocol (MCP) that provides tools to interact with The Movie Database (TMDB).
About TMDB
<img src=https://www.themoviedb.org/assets/2/v4/logos/v2/blue_short-8e7b30f73a4020692ccca9c88bafe5dcb6f8a62a4c6bc55cd9ba82bb2cd95f6c.svg alt="TMDB logo" width="100%"></img>
The Movie Database (TMDB) is a popular, user editable database for movies and TV shows. It provides a comprehensive API that allows developers to access movie and TV show data, including:
- Movie and TV show information
- Cast and crew details
- Images and posters
- Ratings and reviews
- Similar content recommendations
- And much more
API Overview
The TMDB API is free to use but requires an API key. You can get your API key by:
- Creating an account at TMDB
- Going to your account settings
- Selecting the "API" section
- Requesting an API key
Rate Limits
The API has the following rate limits:
- 40 requests per 10 seconds
- 1000 requests per day
For more information about the API, visit the TMDB API Documentation.
Prerequisites
- Node.js (recommended version: 18 or higher)
- pnpm (version 10.7.0 or higher)
Installation
- Clone the repository:
git clone https://github.com/leonardogilrodriguez/mcp-tmdb.git
cd mcp-tmdb
- Install dependencies:
pnpm install
- Create a
.envfile in the project root with your TMDB API key:
TMDB_API_KEY=your_api_key_here
Available Tools
1. two_actors_on_screen
Searches for movies where two actors appear together on screen.
Parameters:
actor1: Name of the first actoractor2: Name of the second actorlanguage: (Optional) Results language (default: "en")
2. two_people
Searches for movies where two people work together in any role.
Parameters:
person1: Name of the first personjob1: Job of the first person. Possible values: cast, crewperson2: Name of the second personjob2: Job of the second person. Possible values: cast, crewlanguage: (Optional) Results language (default: "en")
3. two_movies
Searches for people who worked in two films in any role.
Parameters:
movie1: Name of the first movieyear1: (Optional) Year of the first moviemovie2: Name of the second movieyear2: (Optional) Year of the second movielanguage: (Optional) Results language (default: "en")
4. filmography_actor_genre
Searches for a person's filmography as an actor in a specific genre.
Parameters:
person: Actor's namegenre: (Optional) Genre namelanguage: (Optional) Results language (default: "en")
5. filmography_crew_genre
Searches for a person's filmography as a crew member in a specific genre.
Parameters:
person: Person's namejob: (Optional) Job namegenre: (Optional) Genre namelanguage: (Optional) Results language (default: "en")
6. jobs_list
Provides a list of available jobs to use in the filmography_crew_genre tool.
Usage
To run the MCP server:
pnpm inspector
To run in debug mode:
pnpm dev:debug
Project Structure
mcp-tmdb/
├── API/ # API functions and utilities
├── interfaces/ # Type and interface definitions
├── tools/ # Tool implementations
├── main.ts # Main entry point
├── package.json # Project configuration
└── .env # Environment variables (create)
Main Dependencies
- @modelcontextprotocol/sdk: ^1.12.1
- dotenv: ^16.5.0
- zod: ^3.25.55
MCP Configuration
Claude Desktop
To configure this MCP in Claude Desktop:
- Open Claude Desktop
- Go to Settings > MCP Configuration
- Add a new MCP with the following settings:
"tmdb": {
"disabled": false,
"timeout": 60,
"type": "stdio",
"command": "tsx",
"args": [
"complete_path_to_mcp-tmdb\\main.ts"
],
"env": {
"TMDB_API_KEY": "your_api_key_here"
}
}
Tool Output Examples
1. two_actors_on_screen
Example output when searching for movies with Cary Grant and Katharine Hepburn:
[
{
"id": 31866,
"title": "Sylvia Scarlett",
"year": 1935,
"media_type": "movie"
},
{
"id": 900,
"title": "Bringing Up Baby",
"year": 1938,
"media_type": "movie"
},
{
"id": 16274,
"title": "Holiday",
"year": 1938,
"media_type": "movie"
},
{
"id": 981,
"title": "The Philadelphia Story",
"year": 1940,
"media_type": "movie"
}
]
2. two_people
Example output when searching for collaborations between Christopher Nolan and Hans Zimmer:
[
{
"id": 272,
"title": "Batman Begins",
"year": 2005,
"media_type": "movie"
},
{
"id": 155,
"title": "The Dark Knight",
"year": 2008,
"media_type": "movie"
},
{
"id": 27205,
"title": "Inception",
"year": 2010,
"media_type": "movie"
}
...
]
3. two_movies
Example output when searching for people who worked in two movies: Titanic and The Lord of the Rings: The Return of the King:
[{
"id": 1369,
"name": "Bernard Hill",
"known_for_department": "Acting"
},
{
"id": 1327030,
"name": "Lora Hirschberg",
"job": "Sound Re-Recording Mixer",
"known_for_department": "Sound"
},
{
"id": 900,
"name": "Christopher Boyes",
"job": "Sound Re-Recording Mixer",
"known_for_department": "Sound"
},
{
"id": 1378696,
"name": "Ethan Van der Ryn",
"job": "Sound Effects Editor",
"known_for_department": "Sound"
},
{
"id": 1425978,
"name": "Gary Summers",
"job": "Sound Re-Recording Mixer",
"known_for_department": "Sound"
}
]
4. filmography_actor_genre
Example output when searching for Tom Hanks' horror movies:
[{
"title": "He Knows You're Alone",
"character": "Elliot",
"media_type": "movie",
"year": 1980
},
{
"title": "The 'Burbs",
"character": "Ray Peterson",
"media_type": "movie",
"year": 1989
},
{
"title": "Vault of Horror I",
"character": "Baxter",
"media_type": "movie",
"year": 1994
}
]
4. filmography_crew_genre
Example output when searching for Steven Spielberg's comedy movies as director:
[{
"title": "The Sugarland Express",
"media_type": "movie",
"year": 1974,
"jobs": [
"Director"
]
},
{
"title": "1941",
"media_type": "movie",
"year": 1979,
"jobs": [
"Director"
]
},
{
"title": "Amazing Stories",
"media_type": "tv",
"year": 1985,
"jobs": [
"Director"
]
},
{
"title": "Hook",
"media_type": "movie",
"year": 1991,
"jobs": [
"Director"
]
},
{
"title": "The Terminal",
"media_type": "movie",
"year": 2004,
"jobs": [
"Director"
]
}
]
5. jobs_list
Example output of available jobs:
{
"jobs": [
"Director",
"Producer",
"Screenplay",
"Director of Photography",
"Editor",
"Production Design",
"Art Direction",
"Set Decoration",
"Costume Design",
"Makeup",
"Sound",
"Visual Effects",
"Original Music Composer"
]
}
Demos
two_actors_on_screen
two_movies
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

