MCP Movie Assistant
Enables searching, discovering, and reading movie information from TMDB through MCP tools.
README
MCP Movie Assistant
An MCP server for exploring TMDB movie data. It exposes tools for searching, discovering, and reading movie information over stdio.
Requirements
- Node.js 18 or newer
- npm
- A TMDB credential
Install
npm install
Environment
Create a .env file in the project root.
Supported TMDB variables:
# Preferred for TMDB v4 bearer auth
TMDB_API_KEY=your_tmdb_v4_access_token
# Optional fallback used if no bearer token is present
TMDB_API_KEY_V3=your_tmdb_v3_api_key
# Legacy variable names also supported by the code
API_TOKEN_READ_ONLY=your_tmdb_v4_access_token
API_KEY=your_tmdb_v3_api_key
Notes:
- The server loads
.envon startup. - If both token styles are present, the bearer token is used first.
- Do not commit
.env; it is already ignored by git.
Build
Compile the TypeScript source into build/:
npm run build
Run the server
After building, start the MCP server with:
npm start
This runs node ./build/index.js.
Open the MCP Inspector
The inspector script builds first, then launches the inspector against the compiled server:
npm run inspector
Available tools
The server currently exposes these MCP tools:
search-moviesget-movie-detailsget-trending-moviesget-movie-recommendationsdiscover-movies
Troubleshooting
TMDB request failed with status 401
This usually means the token was not loaded or the wrong variable name is being used.
Check that:
- Your
.envfile exists at the project root - It contains a valid TMDB token or API key
- The value is assigned to one of the supported variables above
- You rebuilt after making code changes with
npm run build
Inspector says Not connected
This usually means the inspector tried to launch build/index.js before the project was built, or the process exited early.
Fix:
npm run build
npm run inspector
Project structure
src/
index.ts # loads env and starts the stdio transport
server.ts # MCP server instance
tools.ts # TMDB-backed tools
api.ts # TMDB fetch helper
build/ # generated output from tsc
Scripts
npm run build- compile TypeScript tobuild/npm start- run the compiled servernpm run inspector- build and open the MCP inspectornpm test- placeholder test script
License
No license has been declared in this repository yet.
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。