transformers

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly shows loading models and datasets from the public Hugging Face Hub (e.g., AutoModel.from_pretrained, datasets.load_dataset("yelp_review_full")) and pipeline inputs that accept external URLs (e.g., pipe("https://example.com/image.jpg")), so the agent can fetch and read untrusted third‑party/user‑generated content.