perplexity

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

• Third-party content exposure detected (medium risk: 0.30). In the ask/direct-tool runtime, the agent ingests outsider-authored free text indirectly as its query/turn/--file content and then sends it to Perplexity WebUI for web research, where returned third-party snippets (attacker-authored web content) are included in the model prompt/response payload.