fix(a2a): fix legacy streaming chunk, agent card test, and metadata merge

- providers/litellm_completion: move 'final' out of the message object
  into the result envelope per the A2A spec (matches the bridge fix).
- agent endpoints test: the runtime invocation path now preserves the
  top-level 'url' on the stored card, so update the assertion to match.
- completion bridge metadata: when forwarding A2A metadata via
  extra_body.metadata, merge into any existing extra_body.metadata
  instead of replacing it, so an agent-configured metadata block is
  preserved (forward metadata still wins on key conflicts).

Co-authored-by: Yassin Kortam <yassin@berri.ai>

litellm/a2a_protocol/litellm_completion_bridge/transformation.py

@@ -100,7 +100,15 @@ class A2ACompletionBridgeTransformation:
         extra_body = completion_params.get("extra_body")
         if not isinstance(extra_body, dict):
             extra_body = {}
-        extra_body = {**extra_body, "metadata": forward_metadata}
+        # Merge into any existing ``extra_body.metadata`` so an
+        # agent-configured ``extra_body: {metadata: {...}}`` is preserved;
+        # forwarded A2A metadata takes precedence on key conflicts.
+        existing_metadata = extra_body.get("metadata")
+        merged_metadata: Dict[str, Any] = (
+            {**existing_metadata} if isinstance(existing_metadata, dict) else {}
+        )
+        merged_metadata.update(forward_metadata)
+        extra_body = {**extra_body, "metadata": merged_metadata}
         completion_params["extra_body"] = extra_body
 
         verbose_logger.debug(

litellm/a2a_protocol/providers/litellm_completion/transformation.py

@@ -266,15 +266,19 @@ class A2ACompletionBridgeTransformation:
         if not content and not is_final:
             return None
 
-        # Build A2A streaming chunk (legacy format)
+        # Build A2A streaming chunk (legacy format). ``final`` is an
+        # envelope-level streaming property per the A2A spec and must live
+        # alongside ``message`` in ``result``, not inside the message object.
         a2a_chunk = {
             "jsonrpc": "2.0",
             "id": request_id,
             "result": {
-                "kind": "message",
-                "role": "agent",
-                "parts": [{"kind": "text", "text": content}],
-                "messageId": uuid4().hex,
+                "message": {
+                    "kind": "message",
+                    "role": "agent",
+                    "parts": [{"kind": "text", "text": content}],
+                    "messageId": uuid4().hex,
+                },
                 "final": is_final,
             },
         }

tests/test_litellm/proxy/agent_endpoints/test_endpoints.py

@@ -414,8 +414,10 @@ class TestAgentRBACProxyAdmin:
             stored_card = call_kwargs["agent"]["agent_card_params"]
             new_agent_id = call_kwargs["agent_id"]
 
-            # Top-level url is dropped; supportedInterfaces points at the proxy.
-            assert "url" not in stored_card
+            # Top-level url is retained for runtime A2A invocation (the public
+            # well-known endpoint rewrites it before exposing to clients);
+            # supportedInterfaces points at the proxy.
+            assert stored_card["url"] == "http://localhost"
             assert stored_card["supportedInterfaces"][0]["protocolBinding"] == "JSONRPC"
             assert stored_card["supportedInterfaces"][0]["url"].endswith(
                 f"/a2a/{new_agent_id}"