syncfusion-javascript-tab

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and rendering of content from external APIs and user-provided inputs as shown in 'references/data-binding.md'. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the data processed by the agent.
  • Ingestion points: Data enters the component context through 'Ajax', 'Fetch', and 'DataManager' calls to external endpoints (e.g., 'api/tabs/content' or OData services) in 'references/data-binding.md'.
  • Boundary markers: The documentation strongly recommends the 'enableHtmlSanitizer' property and provides guidance on using delimiters for user-generated content in 'references/security-html-sanitization.md'.
  • Capability inventory: The skill facilitates web-based data fetching and DOM manipulation; it does not involve subprocess spawning or local file system writes.
  • Sanitization: A built-in HTML sanitizer is enabled by default ('enableHtmlSanitizer: true'), and the documentation provides patterns for additional sanitization using libraries like DOMPurify in 'references/security-html-sanitization.md'.
  • [EXTERNAL_DOWNLOADS]: The component references external assets and data sources for functionality and examples.
  • The documentation recommends fetching Font Awesome icons from Cloudflare's CDN ('cdnjs.cloudflare.com').
  • Sample data binding examples utilize public OData services ('services.odata.org').
  • The installation guide in 'references/getting-started.md' provides instructions to clone official vendor quickstart repositories from GitHub.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 01:31 PM