research

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

• Third-party content exposure detected (low risk: 0.10). Outsider-authored free text is ingested when the workflow performs targeted web scraping/search fetches (e.g., CompetitorMentions uses swan-fetch-scraped-url to pull LinkedIn/Google/G2/TrustRadius/Capterra pages and then LLM-classifies the returned content, and Company/Domain similarly fetch and read homepage/blog/pages) with the user providing the target terms.