brand-voice-analyzer

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

• Third-party content exposure detected (high risk: 1.00). The skill's Phase 1 "Content Collection" explicitly instructs the agent to use WebFetch and WebSearch to pull website copy, blog posts, and social media (public third‑party content) which the agent ingests and uses to drive analysis and generate guides, thereby exposing it to untrusted third‑party input.