cosmic-database

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's examples construct and embed a base64-encoded email:password Authorization header (using hardcoded EMAIL/PASSWORD variables) which requires the agent to include secret values verbatim in requests or generated code.

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md explicitly instructs the agent to fetch and parse data from the public COSMIC REST API (e.g., https://cancer.sanger.ac.uk/cosmic/api used in multiple queries for mutations, genes, samples, signatures, and resistance variants), so the agent ingests external third-party web content that directly influences its analysis and actions.