[QUESTION] Diagram of the multi-agent architecture

[Unit 2.1 Multi-Agent Systems](https://huggingface.co/learn/agents-course/unit2/smolagents/multi_agent_systems#multi-agent-systems) contains [an image](https://mermaid.ink/img/pako:eNp1kc1qhTAQRl9FUiQb8wIpdNO76eKubrmFks1oRg3VSYgjpYjv3lFL_2hnMWQOJwn5sqgmelRWleUSKLAtFs09jqhtoWuYUFfFAa6QA9QDTnpzamheuhxn8pt40-6l13UtS0ddhtQXj6dbR4XUGQg6zEYasTF393KjeSDGnDJKNxzj8I_7hLW5IOSmP9CH9hv_NL-d94d4DVNg84p1EnK4qlIj5hGClySWbadT-6OdsrL02MI8sFOOVkciw8zx8kaNspxnrJQE0fXKtjBMMs3JA-MpgOQwftIE9Bzj14w-cMznI_39E9Z3p0uFoA?type=png) depicting a diagram of the multi-agent architecture. In this image, the Manager Agent, which is typically responsible for task delegation, has direct access to a Code-Interpreter Tool. Would it be more reasonable in practice if there was a Code-Interpreter Agent between them?