use-dom

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a mechanism for native-to-webview communication that could be exploited via indirect prompt injection if processing untrusted data.
  • Ingestion points: External data is ingested through component props such as 'data', 'code', 'content', and 'id' in files like 'components/WebChart.tsx' and 'components/dom-component.tsx'.
  • Boundary markers: The examples lack explicit delimiters or instructions to ignore embedded content within the passed props.
  • Capability inventory: Native functionality is exposed to the webview through callback props like 'showAlert' (native UI alerts) and 'saveData' (persistence operations) as seen in 'app/index.tsx'.
  • Sanitization: No data sanitization or input validation is demonstrated in the provided code snippets for data crossing the native-web bridge.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 10:17 AM