alibabacloud-docmind-parse

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The V2 API request schema explicitly includes ossConfig fields (accessKeyId, accessKeySecret, securityToken) that would require the agent to embed credential values verbatim into the outgoing JSON/request if used, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via the user-supplied input URL (public web content) or downloaded file contents: the script passes document.fileUrl to DocMind (V2) or uploads the file bytes (POP/V2), and the returned markdownContent/text from that outsider document is then printed/serialized (and thus becomes LLM-readable context for downstream steps).