← 返回 Skills 市场
sumo0221

Sumo Torrent

作者 Sumo0221 · GitHub ↗ · v1.0.4 · MIT-0
cross-platform ⚠ suspicious
133
总下载
0
收藏
1
当前安装
5
版本数
在 OpenClaw 中安装
/install sumo-torrent
功能描述
Search BT4G for torrents and add magnet links to qBittorrent. Manual workflow with tracker enrichment. For OpenClaw agents with browser capability.
安全使用建议
Before installing, consider these points: - The skill asks the agent to use a specific browser profile ('my-daily-chrome') to bypass Cloudflare. That profile likely contains session cookies and other private data; grant access only if you trust the profile and understand the privacy implications. - The Python bridge uses hard-coded qBittorrent credentials (admin/adminadmin) and assumes qBittorrent is running on localhost:8080. Change default passwords and avoid exposing the WebUI to the network. If you don't want the skill to control your local qBittorrent, don't provide or allow access to that profile/port. - The SKILL.md and scripts assume Windows paths (C:\butler_sumo\...) but the registry metadata has no OS restriction. Ensure the paths make sense for your environment and that the agent isn't given unnecessary filesystem access. - The skill asks other agent instances to forward commands to a centralized 'master' profile; this centralization can concentrate sensitive access. If multiple agents are in use, be cautious about routing actions through a single privileged profile. - Legal and safety reminder: searching/downloading torrents can have legal risks depending on content and jurisdiction. Verify that your use is lawful. If you decide to use this skill: review and sanitize the scripts (remove hard-coded credentials, confirm output path), avoid sharing personal browser profiles with the agent, and restrict the agent's browser/session access to a disposable profile where possible.
功能分析
Type: OpenClaw Skill Name: sumo-torrent Version: 1.0.4 The skill bundle provides tools to search for torrents and interface with a local qBittorrent instance. It is classified as suspicious due to several high-risk practices: `add_to_qbittorrent.py` contains hardcoded default credentials (admin/adminadmin), and `SKILL.md` explicitly instructs the AI agent to use a specific browser profile ('my-daily-chrome') to bypass Cloudflare, which grants the agent access to the user's personal session cookies and browsing data. While these behaviors are aligned with the stated purpose of the tool, the use of hardcoded paths (C:\butler_sumo) and the instruction to access sensitive browser profiles represent significant security and privacy risks.
能力标签
crypto
能力评估
Purpose & Capability
Name/description match the code: search BT4G and add magnet links to qBittorrent. However, the SKILL.md and code assume a Windows path (C:\butler_sumo\...) and a specific browser profile 'my-daily-chrome' as part of normal operation, yet the skill metadata declares no required config paths, no OS restriction, and no required environment variables. Requiring access to a particular browser profile and local filesystem is disproportionate to the high-level description and is not reflected in the declared requirements.
Instruction Scope
Instructions explicitly tell the agent to open the BT4G site using a profile='my-daily-chrome', take screenshots, click into pages, extract magnet hashes, and hand tasks to a single '總管蘇茉' (master agent/profile). Asking the agent to use a specific personal browser profile implies access to session cookies and other site data beyond the torrent site. The SKILL.md also prescribes writing output to a fixed local path and instructs agents to forward requests to a centralized profile—both expand scope beyond a simple search-and-return workflow.
Install Mechanism
No install spec (instruction-only) which is low-risk. The repository includes two Python scripts that perform network calls (requests) and write local files. No remote downloads or installers are used. The presence of runnable code means the agent will execute local scripts when invoked, which is expected but should be noted.
Credentials
The skill declares no required credentials, but the Python bridge embeds default qBittorrent credentials (admin/adminadmin) and assumes localhost:8080. The SKILL.md expects use of a specific browser profile ('my-daily-chrome')—this effectively requests access to a private browser profile and its cookies without declaring it. The scripts also read/write a fixed Windows directory (C:\butler_sumo...), which is not announced in the skill metadata. These implicit resource/credential requirements are disproportionate and potentially privacy-sensitive.
Persistence & Privilege
always:false and no installation hooks are present. The skill does instruct agents to centralize handling via a 'master' profile, which could increase blast radius if that profile is privileged, but the skill itself does not request permanent inclusion or system-level changes.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install sumo-torrent
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /sumo-torrent 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.4
Updated to 110 trackers from MIDE-690.torrent, improved magnet link generation for YUJ042
v1.0.3
sumo-torrent 1.0.3 - 瀏覽器處理 Cloudflare 流程優化:建議先嘗試使用預設瀏覽器,失敗再用 profile="my-daily-chrome" - 新增「瀏覽器使用順序」說明,避免堅持用同一個 profile - 釐清推薦方法與選擇時機,調整表格說明 - 其他文檔敘述細節優化
v1.0.2
- Major documentation rewrite: SKILL.md now focuses on clear step-by-step instructions and emphasizes browser profile usage to bypass Cloudflare. - Centralized handling: All /torrent commands are to be handled by a designated main agent to avoid session and browser conflicts. - Enhanced tracker explanation: The file now highlights the addition of 22 public trackers and why this is necessary. - Detailed troubleshooting and usage flows: Clearly explained manual and browser-assisted workflows, with guidance for failed cases. - Added safety notes regarding qBittorrent setup and security best practices. - Updated examples and output formats for clarity and practical usage.
v1.0.1
- Added a Chinese documentation file (SKILL_zh.md). - Updated SKILL.md: clarified assisted/manual workflow, improved setup instructions, and emphasized security and environment variable usage. - Removed package.json (no longer required for current setup). - Reduced system requirements: removed unused environment variable. - More concise, user-focused instructions and troubleshooting.
v1.0.0
Initial release of Sumo Torrent skill. - Search BT4G for torrents and add magnet links to qBittorrent WebUI automatically. - Supports OpenClaw browser automation with Cloudflare bypass. - Filters results to active torrents (seeders > 0); corrects info hash extraction to use 40-character hexadecimal. - Automatically enriches magnets with public trackers. - Includes setup and troubleshooting guides for seamless use.
元数据
Slug sumo-torrent
版本 1.0.4
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 5
常见问题

Sumo Torrent 是什么?

Search BT4G for torrents and add magnet links to qBittorrent. Manual workflow with tracker enrichment. For OpenClaw agents with browser capability. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 133 次。

如何安装 Sumo Torrent?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install sumo-torrent」即可一键安装,无需额外配置。

Sumo Torrent 是免费的吗?

是的,Sumo Torrent 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Sumo Torrent 支持哪些平台?

Sumo Torrent 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Sumo Torrent?

由 Sumo0221(@sumo0221)开发并维护,当前版本 v1.0.4。

💬 留言讨论