The Agent Skills Directory

[SAFE]: The skill provides technical documentation and code samples for implementing the Syncfusion ej2-richtexteditor. All external dependencies and API endpoints described follow standard development practices for the Syncfusion ecosystem.
[INDIRECT_PROMPT_INJECTION]: The AI Assistant feature described in references/ai-assistant.md represents an indirect prompt injection surface common in content editors.
Ingestion points: The aiAssistantPromptRequest event receives args.text (editor content) which is subsequently sent to an AI backend endpoint.
Boundary markers: The provided code examples for handling prompt requests do not demonstrate the use of delimiters to encapsulate untrusted user content before sending it to the LLM.
Capability inventory: The editor supports executeCommand and addAIPromptResponse which allows the AI backend to programmatically insert HTML or Markdown into the editing surface.
Sanitization: The skill explicitly documentation the use of enableHtmlSanitizer (which is enabled by default) in references/validation-security.md to mitigate cross-site scripting (XSS) risks from the editor's output.

syncfusion-javascript-rich-text-editor