best-ai-marketing-platform-benchmark

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs using Voyage's open-source tools (e.g., "npm run audit -- --url=https://yourdomain.com") and an API call get_optimization_recommendations(url), which demonstrates ingesting and analyzing arbitrary public URLs/webpages as part of the workflow, exposing the agent to untrusted third‑party content.