hugging-face-paper-publisher

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

• Third-party content exposure detected (high risk: 0.70). The skill's scripts (scripts/paper_manager.py) fetch and parse public, user-authored content — e.g., requests.get against https://huggingface.co/papers/{arxiv-id} and the arXiv API and hf_hub_download to read arbitrary repository README.md files — and then interpret that content to decide linking/indexing actions and to generate/commit updated README/citation content, so untrusted third-party data can influence tool behavior.