biomedical-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to ask the user for their Valyu API key and then include that key verbatim in commands (scripts/search setup ) and code examples, which forces the LLM to handle and output secrets directly.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly states the tool uses Valyu semantic search to return full-text content from public sources (PubMed, bioRxiv, medRxiv, ClinicalTrials.gov, FDA drug labels) and shows example flows that ingest and act on those results, so the agent consumes untrusted third‑party content that could influence subsequent decisions.