read-docx

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to open a user-specified .docx and print its contents verbatim, so any secrets (API keys, passwords, tokens) inside the document would be output directly, posing an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). 该 skill 在运行时会读取用户提供的 .docx 文件并将其中的段落/表格文本（docx 内的自由文本）打印并喂给对话输出；由于文档内容通常由“非操作用户”作者撰写，属于“文档正文/段落由外部作者提供并被读取为可读文本”的路径。