perplexity-cli

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

• Third-party content exposure detected (high risk: 1.00). The skill performs AI-powered searches that ingest results from the open web and social sources (see the --sources web,scholar,social flag and general "AI-powered searches" description), so the agent will read untrusted third‑party/user‑generated content.