The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of AI chat interfaces which process untrusted user input and external AI responses, presenting a surface for indirect prompt injection.
Ingestion points: The Messages property (ObservableCollection) and the PromptRequest event in SKILL.md and references/getting-started.md act as the primary entry points for untrusted data.
Boundary markers: While the control does not enforce strict message delimiters internally, the PromptRequest event provides a mechanism for developers to implement their own validation and boundary logic.
Capability inventory: The skill defines UI components and does not contain scripts that perform file system writes, network exfiltration, or shell command execution.
Sanitization: The documentation in references/theming-and-events.md provides implementation patterns for validating user input and filtering content before it is processed by the AI service.
[SAFE]: The skill references the Syncfusion.Chat.WinUI NuGet package and the syncfusion.com domain, which are official resources belonging to the verified author, Syncfusion Inc. The code examples use safe practices, such as CancellationToken for managing long-running operations and placeholders for sensitive license registration.

syncfusion-winui-ai-assistview