web-search-perplexity

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

• Third-party content exposure detected (high risk: 0.85). At runtime the skill calls Perplexity’s /chat/completions and /search endpoints via the user-provided messages/query, and Perplexity returns synthesized responses that are derived from outsider-authored public web pages (browser-extracted/snippet text) which the agent then ingests into the LLM context.