Add prompt_id propagation in a2a-server task (#14581)

packages/a2a-server/src/agent/task.ts

@@ -68,6 +68,8 @@ export class Task {
   completedToolCalls: CompletedToolCall[];
   skipFinalTrueAfterInlineEdit = false;
   modelInfo?: string;
+  currentPromptId: string | undefined;
+  promptCount = 0;
 
   // For tool waiting logic
   private pendingToolCalls: Map<string, string> = new Map(); //toolCallId --> status
@@ -859,11 +861,10 @@ export class Task {
     };
     // Set task state to working as we are about to call LLM
     this.setTaskStateAndPublishUpdate('working', stateChange);
-    // TODO: Determine what it mean to have, then add a prompt ID.
     yield* this.geminiClient.sendMessageStream(
       llmParts,
       aborted,
-      /*prompt_id*/ '',
+      completedToolCalls[0]?.request.prompt_id ?? '',
     );
   }
 
@@ -893,17 +894,18 @@ export class Task {
     }
 
     if (hasContentForLlm) {
+      this.currentPromptId =
+        this.config.getSessionId() + '########' + this.promptCount++;
       logger.info('[Task] Sending new parts to LLM.');
       const stateChange: StateChange = {
         kind: CoderAgentEvent.StateChangeEvent,
       };
       // Set task state to working as we are about to call LLM
       this.setTaskStateAndPublishUpdate('working', stateChange);
-      // TODO: Determine what it mean to have, then add a prompt ID.
       yield* this.geminiClient.sendMessageStream(
         llmParts,
         aborted,
-        /*prompt_id*/ '',
+        this.currentPromptId,
       );
     } else if (anyConfirmationHandled) {
       logger.info(