The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The component is designed to ingest and display content from external AI service providers and user inputs. This creates a natural surface for indirect prompt injection attacks where malicious instructions could be embedded in the data processed by the agent. The skill includes specific guidance on using MarkupString for rendering and emphasizes the importance of sanitizing input to mitigate XSS risks, as seen in references/templates.md.
[DYNAMIC_EXECUTION]: The component utilizes (MarkupString) to dynamically render responses provided by AI services, which can include HTML, Markdown, and other formatted content. This behavior is standard for rich-text chat interfaces but represents a dynamic rendering surface for untrusted external data.
[EXTERNAL_DOWNLOADS]: The skill documentation references external assets, including Font Awesome and Bootstrap Icons from well-known CDNs (Cloudflare and JSDelivr), and Syncfusion-specific scripts and stylesheets. These are standard resources for building web interfaces and are retrieved from trusted or well-known service providers.

syncfusion-blazor-ai-assistview