rutorrent-mcp
MCP server for controlling rTorrent through ruTorrent's httprpc plugin, enabling torrent management (list/add/remove/start/stop), label/priority management, data movement, and global throttle settings from any MCP client.
README
rutorrent-mcp
MCP server for controlling rTorrent through ruTorrent's httprpc plugin —
list/add/remove/start/stop torrents, manage labels and priorities, move data,
and set global throttles from any MCP client.
Requirements
- Python 3.11+ and uv
- A ruTorrent install with the
httprpcplugin enabled (default on most seedboxes) - Optional plugins:
diskspace(disk stats inglobal_stats),erasedata(needed forremove_torrent(delete_data=true)to actually delete files)
Configuration (environment variables)
| Variable | Required | Default | Description |
|---|---|---|---|
RUTORRENT_URL |
yes | — | Base ruTorrent URL, e.g. https://host/rutorrent |
RUTORRENT_USER |
no | — | HTTP Basic auth username |
RUTORRENT_PASSWORD |
no | — | HTTP Basic auth password |
RUTORRENT_TIMEOUT |
no | 30 |
HTTP timeout in seconds |
RUTORRENT_VERIFY_SSL |
no | true |
Set false to skip TLS verification |
Install with Claude Code
claude mcp add rutorrent \
-e RUTORRENT_URL=https://host/rutorrent \
-e RUTORRENT_USER=myuser \
-e RUTORRENT_PASSWORD=mypass \
-- uv run --directory /path/to/rutorrent_mcp rutorrent-mcp
Tools
| Tool | Description |
|---|---|
list_torrents |
Compact list; filter by label/status, sort, limit |
get_torrent |
Full details; optional files/trackers/peers |
global_stats |
Rates, throttle limits, torrent count, disk space |
add_torrent |
Magnet/URL or local .torrent upload; label/dir/start options |
remove_torrent |
Remove; delete_data=true erases files (erasedata plugin) |
start_torrent / stop_torrent / pause_torrent / recheck_torrent |
Control by hash(es) |
set_label / set_priority |
Organization, bulk-capable |
get_files |
File list with indices, sizes, completion, priorities |
set_file_priority |
Per-file priority: off (skip) / normal / high |
move_torrent |
Move download directory, optionally moving data |
set_throttle |
Global KB/s limits, 0 = unlimited |
rTorrent XML-RPC quirks
Per-file priority is not set over plain XML-RPC: f.set_priority is missing
on some rTorrent builds and f.priority.set needs the HASH:f<index> target
syntax, which varies by version. set_file_priority instead uses httprpc's
form-urlencoded protocol (mode=setprio&hash=…&s=<prio>&v=<index>…) — the
same path the ruTorrent web UI uses — and the PHP handler runs
d.update_priorities afterwards. File priorities are a 3-level scale
(0 = off/skip, 1 = normal, 2 = high), distinct from the 4-level torrent scale.
Development
uv sync
uv run pytest
# manual live check against a real seedbox:
RUTORRENT_URL=... RUTORRENT_USER=... RUTORRENT_PASSWORD=... uv run python scripts/smoke.py
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。