feishu-doc-reader

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

• Third-party content exposure detected (medium risk: 0.30). feishu-doc-reader’s runtime workflow reads untrusted Feishu document content via user-supplied document tokens (e.g., the docx/doc/sheet token from the URL) and extracts the document blocks/text from the Feishu Open API, so outsider authorship within those documents is ingested at runtime.