Matteo Vaccari：智能体编程时代测试应以数据形式表达
写代码交给 Agent 之后，测试怎么跟着改？Matteo Vaccari 给出的答案是测试写成数据而不是代码，方便人审和机器读。做 AI 辅助开发的工程师值得看看这套思路。
Martin Fowler 站点发布 Matteo Vaccari 关于智能体编程下测试策略的文章，核心主张是将测试重心从代码形式转向以数据表达的广覆盖测试。
作者认为当编码工作大量交给 Agent 完成时，人类工程师的职责应转向验证系统行为的正确性，而非维护测试代码本身。这类测试以数据形式存在，便于人审查和 Agent 读取。
文章发布于 martinfowler.com，由 Martin Fowler 在 X 上转发。
Matteo Vaccari claims that with agentic engineering we need to re-orient our testing strategy with broader-scope tests represented as data rather than code. martinfowler.com/articles/tests…📊 3 ⚡
This report is categorized under AITOP's "How-to" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include 测试, 智能体, Agentic Engineering, Martin Fowler.
Source attribution: curated from Martin Fowler (@martinfowler).
