perplexity-data-handling

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s required runtime workflow sends user-provided free text (the search query passed through sanitizeQuery and then into perplexity.chat.completions.create) to the Perplexity Sonar API and parses the resulting answer/citations, so outsider text can directly enter the monitored ingestion path.