perplexity-core-workflow-a

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow sends a user-provided query to Perplexity’s chat.completions.create endpoint (model "sonar") and then ingests the returned web-grounded answer text (response.choices[0].message.content) along with citations/snippets, so outsider-authored free text from the open web can be read at runtime without selecting specific items.