firecrawl-cli

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

• Third-party content exposure detected (high risk: 1.00). SKILL.md explicitly instructs the agent to fetch and ingest public web content via commands like "firecrawl scrape https://example.com", "firecrawl search ... --scrape", and "firecrawl crawl https://example.com", meaning untrusted third‑party webpages and search results are read and can influence subsequent actions.