Qbittorrent Api
Analysis
The skill is mostly a coherent qBittorrent API reference, but it includes an unsafe `.env` sourcing pattern and gives the agent high-impact qBittorrent control using local credentials.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
source .env 2>/dev/null || true
The documented authentication pattern shell-sources the project `.env` file before API use. Sourcing a file executes shell content from that file, which is broader than safely parsing `QB_URL`, `QB_USER`, and `QB_PASS`.
删除种子 | `/delete` | POST | `hashes`, `deleteFiles` ... 关闭 | `/shutdown` | POST
The skill documents raw qBittorrent API operations that can delete torrents and files or shut down the application. These actions fit the stated purpose, but they are high-impact.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
QB_USER="admin"; QB_PASS="123123" ... COOKIE_FILE="/tmp/qb_cookies.txt"
The skill uses qBittorrent credentials and stores an authenticated cookie file. This is expected for the Web API, but it grants control over the configured qBittorrent instance and the registry metadata declares no primary credential.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
优先级顺序:1. 项目目录的 `.env` 文件(默认) 2. 项目 `CLAUDE.md` 文件 3. 询问用户提供
The skill allows qBittorrent connection configuration to come from `CLAUDE.md`, which is also a persistent project instruction/context file.
