biocontext-knowledge-queries
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill performs network operations to fetch data from well-known biomedical services including UniProt, STRING, PubMed, and OpenTargets. These are legitimate scientific databases and no sensitive local data access (such as credentials or system files) was detected.
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves external data such as publication abstracts and full-text articles from PubMed Central, which are incorporated into the agent's context. While this represents a surface for indirect prompt injection, it is the intended primary purpose of the skill and uses standard API access.
- Ingestion points: Functions like
ov.biocontext.search_literatureandov.biocontext.get_fulltextinSKILL.mdandreference.mdingest external content. - Boundary markers: None explicitly defined in the provided snippets.
- Capability inventory: The skill facilitates data retrieval and annotation within the Python environment.
- Sanitization: Content is processed as standard API response data.
Audit Metadata