用 Amazon Bedrock AgentCore 评估多智能体系统的可解释性与有用性
Evaluating multi-agent systems for explainability and helpfulness with Amazon Bedrock AgentCore
AWS 出的实操教程，教你怎么用 Bedrock AgentCore 评估多智能体系统，供应链场景例子很完整，做 Agent 工程的可以照着搭。
AWS 发布教程，介绍如何构建基于 Strands 框架的多智能体供应链决策系统。该教程使用 Amazon Bedrock AgentCore Evaluations 对系统进行评估，涵盖内置评估器、自定义评估器和可解释性评估器三类工具。文章强调多智能体系统的评估标准不止于回复流畅，还需验证工具选择是否正确、是否遵守约束条件，以及决策过程能否被解释。整篇以供应链决策场景为例，给出完整的搭建与评估流程。
Evaluating multi-agent systems for explainability and helpfulness with Amazon Bedrock AgentCore
Multi-agent systems need deeper guarantees than fluent responses: they must select the right tools, respect constraints, and explain their decisions. Learn how to build a Strands-based multi-agent supply chain decisioning system and evaluate it with Amazon Bedrock AgentCore Evaluations using built-in, custom, and explainability evaluators.
