multi-wordcheck

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to insert the user's REDFOX_API_KEY value into shell/PowerShell commands (e.g., export/SetEnvironmentVariable) and to echo the variable for verification, which requires the LLM to handle/output the secret verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). 来源在 SKILL.md/README 中允许用户“粘贴网页链接，系统自动抓取页面正文后检测”，且运行时在 scripts/extract_text.py 的 extract_from_web() 会对用户提供的 URL 进行 requests.get 并提取正文文本，再在 scripts/check_sensitive_words.py 里把该文本发送给检测后端做违禁词检测。