pubmed-fulltext-access

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill fetches scientific content from external databases including Europe PMC, Unpaywall, and CORE, which creates a surface for indirect prompt injection. 1. Ingestion points: Data entering via get_fulltext, get_article_figures, and get_text_mined_terms in SKILL.md. 2. Boundary markers: The skill does not define specific delimiters or instructions to ignore instructions embedded within the retrieved text. 3. Capability inventory: The skill defines tools for data retrieval; no dangerous execution or file-system capabilities are present in the provided context. 4. Sanitization: No sanitization logic is described for the incoming text data.
  • [SAFE]: The skill retrieves data from well-known academic services such as Europe PMC, Unpaywall, and CORE.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 02:34 AM
Security Audit — agent-trust-hub — pubmed-fulltext-access