ks-ai-feed

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs users to supply an API token and gives examples that embed the token verbatim in environment variables, command-line arguments, config files and HTTP headers, which would require the LLM to output or handle secret values directly.

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

• Third-party content exposure detected (medium risk: 0.30). daily_report.py 在运行时会向红狐接口 https://redfox.hk/story/api/parseWork/queryKsAiMsgs/batch 读取返回的快手视频字段（如 title/userName/topic/type/coverUrl 等），并直接拼入生成的 HTML 内容，因此存在被外部作者文本污染的间接提示注入暴露。