syncfusion-angular-inline-ai-assist

Pass

Audited by Gen Agent Trust Hub on Oct 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documentation and examples demonstrate rendering AI-generated responses directly into the DOM using innerHTML or [innerHTML] bindings. This architecture is susceptible to executing malicious scripts if the AI service returns specifically crafted content (e.g., via prompt injection at the source).
  • Ingestion points: Responses received via the promptRequest event and added through the addResponse method in SKILL.md and references/events-and-methods.md.
  • Boundary markers: No specific boundary markers or instructions to isolate the AI output from the rest of the application context were observed.
  • Capability inventory: The component examples in SKILL.md (lines 98, 185) and references/templates-and-toolbars.md (lines 147, 172) utilize direct innerHTML manipulation of application elements.
  • Sanitization: The provided examples lack explicit sanitization or use of safe DOM APIs for handling the dynamic content generated by the AI service.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of packages from Syncfusion's official registries, such as @syncfusion/ej2-angular-interactive-chat and @syncfusion/ej2-material3-theme. Syncfusion is a well-known software component provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 14, 2026, 10:38 AM