assert

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

• Third-party content exposure detected (high risk: 0.80). The skill instructs the agent to fetch content from a public third-party URL (recipe://fullstackrecipes.com/assert or https://fullstackrecipes.com/api/recipes/assert), which the agent would read/interpret at runtime and could contain untrusted user-provided content allowing indirect prompt injection.