pubmed-database
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill interacts with 'eutils.ncbi.nlm.nih.gov', which is the authoritative domain for the National Center for Biotechnology Information. These network requests are used solely to fetch scientific data and metadata as described in the skill's documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests article abstracts and metadata from external sources, which constitutes a potential surface for indirect prompt injection.
- Ingestion points: Article abstracts and summaries fetched via E-utilities endpoints (SKILL.md).
- Boundary markers: Data is received in structured JSON and XML formats.
- Capability inventory: The skill uses the 'requests' library to perform GET requests to PubMed.
- Sanitization: Data is parsed using standard libraries (ElementTree and JSON), though the natural language content of the abstracts is not sanitized before being processed by the agent context.
Audit Metadata