biorxiv-database
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches metadata and PDF files from 'api.biorxiv.org', 'www.biorxiv.org', and 'www.ebi.ac.uk'. These are established academic and scientific service domains used for legitimate research retrieval.
- [COMMAND_EXECUTION]: The skill provides instructions for installing common dependencies, specifically 'requests' and 'pandas', using the standard 'pip' package manager.
- [INDIRECT_PROMPT_INJECTION]: The skill processes research metadata, such as titles and abstracts, ingested from external preprint APIs. This represents a potential attack surface if the external data were to contain malicious instructions, which is a standard consideration for tools performing automated data retrieval.
- Ingestion points: The skill ingests JSON responses and PDF content from bioRxiv and EuropePMC APIs.
- Capability inventory: The skill performs network GET requests and writes retrieved data to local CSV and PDF files.
- Boundary markers: No specific delimiting or isolation markers for the external content are specified.
- Sanitization: The provided examples perform basic string slicing for display purposes but do not include complex validation or sanitization of the research data.
Audit Metadata