crisis-communications

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and monitors user-generated social media and public mentions — see "Social media crisis response" and the CrisisMonitor/CrisisMention classes (fields like platform, content, author) and checklist items "Monitor official accounts" and "Social media claims flagged for verification" — so the agent is expected to read/interpret untrusted third‑party content.