azure-compute-batch-java

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

• Third-party content exposure detected (high risk: 0.80). The skill's examples and complete application read and return arbitrary task-produced files (e.g., client.getTaskFile("myJobId", taskId, "stdout.txt") in the Complete Application Example) and show tasks pulling ResourceFile HTTP URLs (e.g., https://storage.blob.core.windows.net/...), which clearly ingest untrusted/user-provided third‑party content that the agent reads as part of its workflow.