colleague-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt asks users to supply app_id/app_secret, OAuth codes and access tokens and shows examples that embed those secrets verbatim in HTTP headers and CLI arguments (e.g. Authorization: Bearer {user_access_token}, --user-token {user_access_token}), which requires the LLM to handle/output secret values directly.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md 的 Step 2/方式 A/B/C/D/E 会将飞书/钉钉消息与文档、以及用户上传/粘贴的原材料文本通过 Read 读入并送入 Step 3（work_analyzer/persona_analyzer），其中消息/文档内容可包含第三方作者的自由文本（间接提示注入）。