getAIRequest
Returns a request configuration object that can be sent to AI services for processing natural language prompts. This method generates a standardized request format that includes the user's prompt and column metadata, which AI services can use to generate appropriate Grid commands.
Use this method when implementing custom AI integrations or when you need to send Grid context to an AI service endpoint. The returned object can be serialized to JSON and sent directly to compatible AI services.
Parameters
prompt String
The user's natural language prompt describing the desired Grid operation (e.g., "sort by name ascending", "filter rows where age is greater than 30", "highlight all rows from New York").
Returns
Object A request configuration object with the following structure:
role- String, always set to "user"contents- Array containing a single object with:$type- String, always set to "text"text- String, the user's prompt
columns- Array of column configuration objects, each containing:id- String, the column's unique identifier (uid)field- String, the column's data field name (if applicable)title- String, the column's display title (if specified)type- String, special column type: "checkbox" for selectable columns, "command" for command columns, "draggable" for draggable columns (only present for special columns)- Additional column properties as configured in the Grid
Example - send prompt to AI service using getAIRequest
<div id="grid"></div>
<script>
$("#grid").kendoGrid({
columns: [
{ field: "name", title: "Name" },
{ field: "age", title: "Age" },
{ field: "city", title: "City" }
],
dataSource: [
{ name: "Jane Doe", age: 30, city: "New York" },
{ name: "John Doe", age: 33, city: "London" },
{ name: "Sam Smith", age: 25, city: "Paris" }
],
sortable: true,
filterable: true,
toolbar: [{
template: '<input id="aiPrompt" class="k-input k-input-md k-rounded-md" placeholder="Ask AI..." style="width: 300px; margin-right: 10px;" />' +
'<button id="askAI" class="k-button k-button-md k-rounded-md k-button-solid k-button-solid-primary">Apply</button>'
}]
});
var grid = $("#grid").data("kendoGrid");
$("#askAI").on("click", function() {
var prompt = $("#aiPrompt").val();
if (!prompt) {
return;
}
// Get the AI request object with prompt and column metadata
var aiRequest = grid.getAIRequest(prompt);
$.ajax({
url: "https://sitdemos.telerik.com/service/v2/ai/grid/smart-state",
method: "POST",
contentType: "application/json",
data: JSON.stringify(aiRequest),
success: function(response) {
// Process the AI response to apply commands to the Grid
grid.handleAIResponse(response);
$("#aiPrompt").val("");
},
error: function(xhr) {
console.log("AI service error:", xhr.responseText);
}
});
});
</script>In this article