typst
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The skill explicitly references and uses external, public package sources and repos (e.g., "Import packages (check https://typst.app/universe ...)" and example imports like #import "@preview/…", plus git/npx install URLs to GitHub), meaning the agent may fetch and compile arbitrary third‑party Typst packages and repository content which it will read/interpret as part of its workflow.
Audit Metadata