crw-scrape

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

• Third-party content exposure detected (high risk: 0.95). Outsider free text can be ingested because the skill fetches arbitrary public web pages at runtime from the provided url (e.g., crw scrape "https://example.com" / MCP crw_scrape(url=...) / REST /v1/scrape), and the page’s readable content is returned to the agent/LLM context as markdown/html/text.