ai-for-science-boltzgen

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly downloads model weights and data from HuggingFace via huggingface_hub.hf_hub_download (see SKILL.md modifications to get_artifact_path) and clones public GitHub repos (PyG sources), which are untrusted third-party sources that are ingested at runtime and can materially change model behavior and decisions.