Ask HN: 编程代理是否优化了错误的审查环节？
1 分•作者: wsxiaoys•6 个月前
这只是个人观点，但我认为目前的编码代理在错误的时机需要人类审查 AI 的产物。
大多数工具侧重于在执行前创建和审查计划。所以，背后的想法是在让代理触及代码库之前批准意图。这听起来合理，但实际上，真正的学习并非发生于此。
“计划模式”发生在代理付出“现实”代价之前。在它浏览代码库之前，在它运行测试之前，在它遇到奇怪的边缘情况或依赖问题之前。其输出是推测性的，并且通常看起来比实际情况自信得多。
实际上，更有用的是审查“演练”：代理尝试解决问题后所做事情的总结。
目前，在大多数编码代理中，默认设置仍然将计划视为主要检查点，而演练则在稍后进行。这使得重心放错了地方。
我从事软件工程的经验是，我们不审查意图，而是信任执行。我们审查结果：差异、测试更改、哪些被破坏、哪些被修复以及原因。这实际上就是一次演练。
因此，我觉得，当我们对演练提供反馈时，我们是对具体决策和后果做出反应，而不是基于假设。这种反馈更清晰、更具可操作性，并且更接近于我们作为工程师今天已经审查工作的方式。
好奇其他人在使用“计划优先”的编码代理时是否有同样的感觉。原因是，我正在开发一个开源编码代理，并决定减少对预先批准计划的重视，而更多地关注审查代理在实际工作中遇到的情况。
但这是我们团队内部正在激烈辩论的事情，很希望得到大家的想法，以便帮助我们以最佳方式实现这一点。
查看原文
This is a personal opinion, but I think current coding agents requires human reviews AI's artficats at the wrong moment.
Most tools focus on creating and reviewing the plan before execution.<p>So the idea behind this is to approve intent before letting the agent touch the codebase. That sounds reasonable, but in practice, it’s not where the real learning happens.<p>The "plan mode" takes place before the agent has paid the cost of reality. Before it’s navigated the repo, before it’s run tests, before it’s hit weird edge cases or dependency issues. The output is speculative by design, and it usually looks far more confident than it should.<p>What will actually turn out to be more useful is reviewing the walkthrough: a summary of what the agent did after it tried to solve the problem.<p>Currently, in most coding agents, the default still treats the plan as the primary checkpoint and the walkthrough comes later. That puts the center of gravity in the wrong place.<p>My experience with SWE is that we don’t review intent and trust execution. We review outcomes: the diff, the test changes, what broke, what was fixed, and why. That’s effectively a walkthrough.<p>So I feel when we give feedback on a walkthrough, we’re reacting to concrete decisions and consequences, and not something based on hypotheticals. This feedback is clearer, more actionable, and closer to how we, as engineers, already review work today.
Curious if others feel the same when using plan-first coding agents. The reason is that I’m working on an open source coding agent, and have decided to keep less emphasis on approving plans upfront and more emphasis on reviewing what the agent actually experienced while doing the work.<p>But this is something we’re heavily debating internally inside our team, and would love to have thoughts so that it can help us implement this in the best way possible.