firecrawl

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

• Third-party content exposure detected (high risk: 0.85). Firecrawl CLI’s runtime workflow for search --scrape, scrape <url>, map, crawl, and monitor fetches outsider-authored public web page content and returns it as readable markdown/JSON (untrusted third-party text) that can be ingested into the agent’s LLM context.