hugging-face-papers

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches paper content and structured metadata from Hugging Face's official API (huggingface.co) and arXiv (arxiv.org), which are recognized sources for research data.\n- [COMMAND_EXECUTION]: Uses shell-based curl commands to perform GET and POST requests for paper indexing, search, and link updates.\n- [CREDENTIALS_UNSAFE]: Appropriately utilizes the $HF_TOKEN environment variable for authentication, which is a standard security practice for managing API tokens without hardcoding them.\n- [PROMPT_INJECTION]: The skill processes untrusted external content from research papers, which is a potential surface for indirect prompt injection.\n
  • Ingestion points: External text is retrieved from paper URLs (e.g., huggingface.co/papers/{PAPER_ID}.md).\n
  • Boundary markers: The instructions do not define specific delimiters to isolate paper content from agent instructions.\n
  • Capability inventory: The agent can perform network requests and text processing based on the fetched content.\n
  • Sanitization: No validation or sanitization is applied to the retrieved paper text before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 11:09 PM