huggingface-vision-trainer

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s required runtime workflow runs scripts/dataset_inspector.py, which fetches and prints user-selected dataset row/field values from Hugging Face Datasets Server via the URL parameters dataset/split, thereby ingesting outsider-authored free text contained in dataset fields (e.g., JSON prompt/labels) before any specific item selection.