skills/syncfusion/aspnetmvc-ui-components-skills/syncfusion-aspnetmvc-ai-assistview/Gen Agent Trust Hub
syncfusion-aspnetmvc-ai-assistview
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents several features that create an attack surface for indirect prompt injection when processing untrusted inputs.
- Ingestion points: The component ingests data from user-submitted prompts (
PromptRequestevent inSKILL.md), file attachments (references/file-attachments.md), and file-aware analysis using the Model Context Protocol (MCP) server integration (references/ai-integrations.md). - Boundary markers: The provided examples do not explicitly define boundary markers or instruct the AI to ignore embedded instructions within processed files or attachments.
- Capability inventory: The component supports file system operations (upload/remove), dynamic Generative UI block rendering, and Markdown rendering.
- Sanitization: The integration guides utilize
marked.jsfor Markdown rendering but do not specify or enforce sanitization settings, which could lead to Cross-Site Scripting (XSS) if the LLM output is influenced to include malicious HTML payloads. - [EXTERNAL_DOWNLOADS]: The skill references various external packages and scripts from established technology providers.
- Syncfusion Infrastructure: Downloads stylesheets and script bundles from Syncfusion's official CDN (
cdn.syncfusion.com). - Official AI SDKs: Provides instructions for installing official NuGet packages from Microsoft (Azure OpenAI, Ollama), Google (Gemini), and the OpenAI project.
- Well-Known CDNs: References the JSDelivr CDN to load the
marked.jslibrary. - Common Package Registries: Includes instructions for installing packages from NPM and PyPI, such as
express,cors, andlitellm.
Audit Metadata