extract

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

• Third-party content exposure detected (high risk: 1.00). This skill accepts arbitrary URLs in its JSON input and calls Tavily's extraction endpoint (see scripts/extract.sh and the POST https://api.tavily.com/extract examples) to fetch and return page raw_content, so it clearly ingests untrusted public web content that could contain indirect prompt-injection.