webcrawler

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

• Third-party content exposure detected (high risk: 0.90). The skill's core scripts (e.g., scripts/crawl_docs.py and scripts/extract_page.py) explicitly fetch and recursively crawl arbitrary public URLs from the open web, parse and extract untrusted page HTML (using requests + BeautifulSoup), and then interpret that content to decide relevance, queue further crawling, and produce outputs that are used in downstream RAG/memory flows—so third‑party content can directly influence agent decisions and actions.