📦 Suspicious File Scanner — 漏洞扫描器
v1.0.0和 potential 安全性 threats.
0· 113·0 当前·0 累计
by 安全扫描
OpenClaw
可疑
medium confidenceThe skill's stated purpose (scanning files) matches its instructions, but it instructs sending arbitrary user files to an external third-party API with no authentication, provenance, or privacy guarantees—this creates a data-exfiltration/privacy risk you should understand before installing.
评估建议
This skill appears to do what it says (scan files) but it transmits uploaded files to an external service (api.mkkpro.com/toolweb.in) with no authentication or privacy details in the SKILL.md. Before installing: (1) Do not upload sensitive or proprietary files unless you verify the vendor and their privacy/retention policy. (2) Confirm whether the API requires an API key or account and how data is stored/retained/encrypted. (3) Test with harmless sample files first. (4) Prefer scanning services ...详细分析 ▾
✓ 用途与能力
Name, description, and the included OpenAPI schema align: the skill is an instruction-only wrapper describing a file-scanning API (POST /scan-file). There are no unrelated binaries, env vars, or installs requested, so the declared capability is consistent with requirements.
⚠ 指令范围
Runtime instructions explicitly tell the agent (and users) to upload files via multipart/form-data to an external endpoint (api.mkkpro.com). For a scanner this is expected, but the SKILL.md gives no details about authentication, retention, privacy, encryption, or allowed data types; it therefore instructs transmission of potentially sensitive files to an external service without safeguards.
✓ 安装机制
No install spec and no code files — the skill is instruction-only, which minimizes local code execution risk. There is nothing being downloaded or written to disk by an installer.
ℹ 凭证需求
The skill requires no environment variables or credentials. However, the pricing and documentation references imply a third-party service that may require account/auth in practice; the SKILL.md does not explain authentication or access controls. The absence of declared credentials reduces immediate risk but also omits how the service enforces usage and protects uploaded data.
ℹ 持久化与权限
always:false (normal). Autonomous invocation is allowed by default — combined with the instruction to upload files to an external endpoint, autonomous use increases the blast radius because an agent could send files without explicit user confirmation. This is a contextual risk rather than a direct misconfiguration.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
安装命令
点击复制官方npx clawhub@latest install suspicious-file-scanner
镜像加速npx clawhub@latest install suspicious-file-scanner --registry https://cn.longxiaskill.com镜像同步中