mrcarlsama-social-transcriber

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.72). 运行时会通过 scripts/providers/yt_dlp_provider.py 调用 yt-dlp --dump-single-json 获取平台返回的 title/description/caption/full_description/content 等字段，并在 scripts/run_one.py 的 write_body_file() 中把这些“平台正文/描述(body)”写入 【标题】正文.md（属于外部作者的自由文本，经由 provider→本地文件→后续可能进入模型润色上下文）。

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime/setup instructions include executing remote installer scripts (curl -LsSf https://astral.sh/uv/install.sh | sh and irm https://astral.sh/uv/install.ps1 | iex), which fetch and execute code from astral.sh during runtime, so this is a required external dependency that runs remote code.