cardiology-content-repurposer
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external source material including YouTube transcripts, medical newsletters, and PDFs provided by users.
- Ingestion points: The workflow in
SKILL.mdexplicitly ingests external content for analysis and transformation. - Boundary markers: There are no specific instructions or delimiters defined to isolate the source material from the agent's core logic or to ignore potential commands embedded in the input.
- Capability inventory: The skill is configured to use
PubMed:search_articlesandPubMed:get_article_metadatafor automated citation searching based on the ingested content. - Sanitization: The instructions do not specify any validation or sanitization steps for the provided source material.
Audit Metadata