pubmed-database
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates data retrieval from the official National Center for Biotechnology Information (NCBI) E-utilities REST API at
eutils.ncbi.nlm.nih.gov. This is a well-known and trusted source for biomedical literature. - [COMMAND_EXECUTION]: Provides Python code templates using the
requestslibrary to execute HTTP queries against the PubMed database for automated literature analysis. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it processes external data retrieved from the PubMed database (such as article abstracts and titles).
- Ingestion points: Data enters the agent's context through API endpoints including
esearch.fcgiandefetch.fcgias described inreferences/api_reference.md. - Boundary markers: There are no explicit delimiters or instruction-isolation markers provided in the example implementation logic.
- Capability inventory: The skill is capable of network operations via
requests.getto communicate with the NCBI infrastructure. - Sanitization: The provided examples do not include specific sanitization steps for the retrieved text, which is typical for skills intended to process scientific literature.
Audit Metadata