markitdown

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and converts arbitrary web pages and public content (e.g., md.convert(url) examples in references/web_content.md, YouTube transcript extraction, web-scraping examples, and scripts/convert_webpage.py), which are untrusted, user-generated third-party sources that the agent ingests and interprets—allowing indirect prompt injection.