agently-mail

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). Insecure: the skill explicitly requires extracting and verbatim forwarding of opaque URLs (the OAuth authorization URL and download_url) from CLI output, which may contain sensitive tokens or presigned credentials, so the LLM must handle/output secret-like values exactly as returned.

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

• Third-party content exposure detected (high risk: 0.85). 运行时会调用 agently-cli message +read/+search 获取邮件“标题/正文/附件元信息”，这些邮件正文与内容属于外部第三方（邮件发送者）自由文本，并会被作为可读文本进入代理上下文。

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)