brand-style-guide

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow and maintenance playbook explicitly require scanning and "stress-test[ing]" real production touchpoints (e.g., "web pages, ads, emails, social posts") to find examples and surface violations, which means the agent is expected to ingest and act on untrusted public third-party content.