transformers

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly directs loading models from the public Hugging Face Hub (e.g., AutoModel.from_pretrained("model-id") and the token link https://huggingface.co/settings/tokens), which are user-uploaded, untrusted third‑party artifacts that the agent will fetch and execute as part of its workflow, exposing it to indirect prompt-injection via model behavior.