huawei-cloud-modelarts-training-diagnosis

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

• Third-party content exposure detected (medium risk: 0.30). In Phase 3 of the workflow, the skill ingests free-text from ModelArts training job fields such as status.task_statuses[].message (tracebacks/error messages) via ShowTrainingJobDetails and, conditionally, log/error content via ShowTrainingJobLogsPreview/ListTrainingJobEvents, which ultimately originates from outsider-authored training code/log output.