TorrentClaw-MCP
Giving AI assistants the ability to search movies and TV shows, find torrents with magnet links, check streaming availability, and explore cast/crew metadata.
README
torrentclaw-mcp
Model Context Protocol server for TorrentClaw — giving AI assistants the ability to search movies and TV shows, find torrents with magnet links, check streaming availability, and explore cast/crew metadata.
torrentclaw-mcp is developed by TorrentClaw as part of its open-source ecosystem. It wraps the TorrentClaw API into the MCP standard so that any compatible AI assistant (Claude, GPT, etc.) can access the platform's search and discovery features natively.
Quick Start
npx torrentclaw-mcp
No API key required (optional for higher rate limits).
What can it do?
Tools
| Tool | Description |
|---|---|
search_content |
Search movies/shows with filters (query, type, genre, year, rating, quality, language, audio, HDR, season, episode, sort). Returns torrents, magnet links, and optional streaming info. |
autocomplete |
Type-ahead search suggestions (up to 8 results). Use to validate titles before a full search. |
get_popular |
Get popular content ranked by user clicks |
get_recent |
Get recently added content |
get_watch_providers |
Streaming availability by country (Netflix, Disney+, etc.) |
get_credits |
Cast and director for a title |
get_torrent_url |
Get .torrent file download URL from info hash |
track_interaction |
Track user interaction with a torrent (magnet click, download, copy) |
submit_scan_request |
Submit a torrent for audio/video quality analysis via TrueSpec |
get_scan_status |
Check the status of a torrent scan request |
Resources
| URI | Description |
|---|---|
torrentclaw://stats |
Catalog statistics (content/torrent counts by source) |
Prompts
| Prompt | Description |
|---|---|
search_movie |
Search for a movie by title and get torrents + streaming |
search_show |
Search for a TV show by title and get torrents |
whats_new |
Discover recently added movies and TV shows |
where_to_watch |
Find where to stream, rent, or buy a title |
Configuration
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"torrentclaw": {
"command": "npx",
"args": ["-y", "torrentclaw-mcp"]
}
}
}
Claude Code
Add to .mcp.json or ~/.claude/settings.json:
{
"mcpServers": {
"torrentclaw": {
"command": "npx",
"args": ["-y", "torrentclaw-mcp"]
}
}
}
Environment Variables
| Variable | Default | Description |
|---|---|---|
TORRENTCLAW_API_URL |
https://torrentclaw.com |
Base URL of the TorrentClaw API |
TORRENTCLAW_API_KEY |
(none) | Optional API key for authenticated access (higher rate limits) |
Project Structure
.
├── src/
│ ├── index.ts # MCP server entry point
│ ├── config.ts # Configuration & URL validation
│ ├── api-client.ts # TorrentClaw API client with caching
│ ├── types.ts # TypeScript interfaces for API responses
│ ├── prompts.ts # MCP prompt definitions
│ ├── tools/ # MCP tool implementations
│ │ ├── search-content.ts
│ │ ├── autocomplete.ts
│ │ ├── get-popular.ts
│ │ ├── get-recent.ts
│ │ ├── get-watch-providers.ts
│ │ ├── get-credits.ts
│ │ ├── get-torrent-url.ts
│ │ ├── track-interaction.ts
│ │ └── scan-request.ts
│ ├── formatters/ # Output formatting
│ │ ├── content.ts
│ │ ├── providers.ts
│ │ └── credits.ts
│ └── resources/ # MCP resources
│ └── stats.ts
├── tests/ # Test suite (vitest)
├── .github/workflows/ # CI/CD (lint, build, test, release)
├── lefthook.yml # Git hooks (commit lint, prettier, tsc)
├── Makefile # Dev workflow (build, test, lint, fmt)
├── CONTRIBUTING.md # Contribution guidelines
├── LICENSE # MIT
└── README.md
Development
git clone https://github.com/torrentclaw/torrentclaw-mcp.git
cd torrentclaw-mcp
make install-tools
make hooks
make build && make test
Test with MCP Inspector:
npx @modelcontextprotocol/inspector node build/index.js
About TorrentClaw
TorrentClaw is an open platform focused on improving the quality and reliability of torrent metadata. Our mission is to make torrent search engines more accurate and the torrent ecosystem healthier — by building tools that verify, enrich, and standardize metadata across the network.
torrentclaw-mcp is part of the TorrentClaw open-source ecosystem, alongside TrueSpec (torrent metadata verification).
Contributing
See CONTRIBUTING.md for guidelines.
License
MIT License — see LICENSE for details.
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。