firecrawl-hardened

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly calls Firecrawl endpoints to fetch arbitrary public web content (see references/api.md: POST /v1/search, /v1/scrape, /v1/crawl) and the provided scripts (scripts/search.py, scripts/scrape.py, scripts/crawl.py) and SKILL.md workflow show the agent reading and previewing scraped pages, so untrusted third‑party content can directly influence the agent's outputs and actions.