markitdown
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill explicitly accepts and fetches arbitrary URLs (see SKILL.md: "markitdown https://example.com" and "Can fetch URLs but Jina is faster"), meaning it ingests open/public web content that could be untrusted and interpreted by the agent.
Audit Metadata