feishu-to-linear

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow explicitly opens and reads user-provided Feishu/Lark document URLs using the Chrome MCP ("使用 Chrome MCP 打开并读取飞书文档"), and it must interpret that untrusted, user-generated document content to decide project/issue creation and scheduling in Linear, which could allow indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill opens and reads a user-provided Feishu document URL at runtime (the Feishu doc URL provided by the user, e.g. https://www.feishu.cn/...) and injects that fetched document content into the agent's context to drive creation of Linear projects/issues, so the external URL's content directly controls the agent's behavior.