feat(mcp): preserve CallToolResult.isError flag in MCPToolResult

src/strands/tools/mcp/mcp_client.py

@@ -744,6 +744,8 @@ def _handle_tool_result(self, tool_use_id: str, call_tool_result: MCPCallToolRes
             result["structuredContent"] = call_tool_result.structuredContent
         if call_tool_result.meta:
             result["metadata"] = call_tool_result.meta
+        if call_tool_result.isError:
+            result["isError"] = True
 
         return result
 

src/strands/tools/mcp/mcp_types.py

@@ -61,7 +61,13 @@ class MCPToolResult(ToolResult):
         metadata: Optional arbitrary metadata returned by the MCP tool. This field allows
             MCP servers to attach custom metadata to tool results (e.g., token usage,
             performance metrics, or business-specific tracking information).
+        isError: Whether the MCP tool reported an application-level error via
+            ``CallToolResult.isError``. ``True`` means the tool executed but its logic
+            returned a failure. Absent when the tool succeeded or when the error was a
+            protocol/client exception rather than a tool-reported failure, letting
+            callers distinguish application errors from transport/protocol errors.
     """
 
     structuredContent: NotRequired[dict[str, Any]]
     metadata: NotRequired[dict[str, Any]]
+    isError: NotRequired[bool]

tests/strands/tools/mcp/test_mcp_client.py

@@ -132,6 +132,8 @@ def test_call_tool_sync_status(mock_transport, mock_session, is_error, expected_
         assert result["content"][0]["text"] == "Test message"
         # No structured content should be present when not provided by MCP
         assert result.get("structuredContent") is None
+        # isError flag should be True only when the tool reported an application error
+        assert result.get("isError") == (True if is_error else None)
 
 
 def test_call_tool_sync_session_not_active():
@@ -261,6 +263,8 @@ async def mock_awaitable():
         assert result["toolUseId"] == "test-123"
         assert len(result["content"]) == 1
         assert result["content"][0]["text"] == "Test message"
+        # isError flag should be True only when the tool reported an application error
+        assert result.get("isError") == (True if is_error else None)
 
 
 @pytest.mark.asyncio
@@ -408,6 +412,15 @@ def test_mcp_tool_result_type():
 
     assert result_with_structured["structuredContent"] == {"key": "value"}
 
+    # isError is optional — absent by default
+    assert "isError" not in result or result.get("isError") is None
+
+    # isError can be set to flag tool-reported application errors
+    result_with_is_error = MCPToolResult(
+        status="error", toolUseId="test-789", content=[{"text": "Tool failed"}], isError=True
+    )
+    assert result_with_is_error["isError"] is True
+
 
 def test_call_tool_sync_without_structured_content(mock_transport, mock_session):
     """Test that call_tool_sync works correctly when no structured content is provided."""