huggingface-vision-trainer

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text when it fetches and prints dataset row fields via the Datasets Server API in scripts/dataset_inspector.py (e.g., parsing JSON prompt strings/label fields from user-specified datasets and rendering them in the inspector output), which is then used as LLM-readable context for compatibility decisions.