AgentCore Evaluations 评测手册：内置、DeepEval 与 AutoEval 三种方案
The agent evaluation playbook: Built-in, DeepEval, and AutoEval on Amazon Bedrock AgentCore Evaluations
AWS 这篇手册把智能体评测拆成了任务完成、跑题、数据泄露这些具体维度，教你怎么在 AgentCore 上选对评测器。写智能体的开发者可以直接照着搭回归测试，不用再靠肉眼验收。
AWS 在博客中给出了一套针对 Amazon Bedrock AgentCore Evaluations 的智能体评测手册，覆盖内置评测器、DeepEval 和 AutoEval 三种路径。文章指出，判断一个智能体是否正常工作实际上是多个问题：任务是否完成、是否跑题、是否泄露数据、是否跳过必要步骤。
技术上，作者强调要把每类关键行为映射到合适的评测器：任务完成度、主题相关性、数据泄露防护、步骤遵循等维度各有对应的指标和评测层级。DeepEval 与 AutoEval 可与 AgentCore 内置评测器组合使用，形成 CI 流程中的自动化回归测试。
对部署智能体的开发团队而言，这类评测框架能把“感觉能用”变成可量化的指标，与 LangSmith、Braintrust 等评测工具形成互补，适合在生产环境上线前建立评测基线。
The agent evaluation playbook: Built-in, DeepEval, and AutoEval on Amazon Bedrock AgentCore Evaluations
When you deploy an agent, whether it is working is really several questions: did it finish the task, stay on topic, leak data, or skip steps? This post shows how to evaluate agents on Amazon Bedrock AgentCore with built-in, DeepEval, and AutoEval evaluators, and how to match each behavior that matters to the right evaluator at the right level.
This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Amazon Bedrock, AgentCore, DeepEval, 智能体.
Primary source verification: published directly by AWS Machine Learning Blog with first-party authority.