biomedical-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill prompt explicitly instructs the agent to ask the user for their API key and then pass it as a direct command-line argument (scripts/search setup <api-key>), causing the secret to pass through the LLM context and command execution.

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

• Third-party content exposure detected (low risk: 0.10). The agent uses a semantic search tool (Valyu) to search external biomedical databases like PubMed, preprints, and clinical trials where search results contain outsider-authored content.