syncfusion-javascript-ai-assistview

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads essential UI assets from Syncfusion's official CDN.\n
  • Evidence: References https://cdn.syncfusion.com/ej2/32.1.19/fluent2.css for component styling.\n- [INDIRECT_PROMPT_INJECTION]: As a chat interface component, the skill is designed to handle untrusted data from multiple ingestion points.\n
  • Ingestion points: User prompts, attached files, and responses received from external AI services.\n
  • Boundary markers: The framework allows developers to implement their own prompting strategies, though the examples focus on basic connectivity.\n
  • Capability inventory: Includes network operations (fetch for AI APIs), file management (saveUrl/removeUrl), and dynamic UI rendering.\n
  • Sanitization: Examples demonstrate the use of the marked library for markdown-to-HTML conversion, requiring developers to ensure safe rendering of AI-provided content.\n- [DATA_EXFILTRATION]: Provides file attachment capabilities with configurable upload destinations.\n
  • Evidence: attachmentSettings defines endpoints for file persistence. Documentation examples use Syncfusion's production API domains (services.syncfusion.com). Per the [TRUST-SCOPE-RULE], these vendor-owned endpoints are documented neutrally.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 08:48 PM