medical-newsletter-writer
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from user-uploaded PDF files in Phase 1 of the workflow. * Ingestion points: The skill reads and extracts topic ideas from user-provided PDF files in SKILL.md. * Boundary markers: There are no boundary markers or instructions for the agent to disregard potential prompt injection attempts hidden in the PDF text. * Capability inventory: The extracted data is used to perform subsequent research tasks using web_search, PubMed, and medical news searches. * Sanitization: The skill lacks sanitization or validation logic to filter out malicious instructions from the extracted PDF data.
Audit Metadata