如何构建一个代理

While seemingly every company is talking about building agents this year, far fewer have done it. It’s easy to let your imagination run wild with how agents can transform your business, but many teams are unsure where to begin, how to make progress, and where to set expectations.

虽然似乎每家公司今年都在谈论构建代理，但实际上做到这一点的公司要少得多。让您的想象力在代理如何改变您的业务方面自由驰骋是很容易的，但许多团队不确定从哪里开始，如何取得进展，以及在哪里设定期望。

In this guide, we’ll walk through a framework for going from idea to impact— illustrated with a real-world example of building an email agent.

在本指南中，我们将通过一个框架，从想法到影响——以构建电子邮件代理的真实世界示例进行说明。

Step-by-step diagram for process of building an agent

构建代理的逐步流程图

Step 1: Define your agent’s job with examples

步骤1：用示例定义代理的工作

Choose something realistic and something that requires an agent.

选择一些现实的东西，并且需要代理的东西。

Pick something you could teach a smart intern. If your best intern could never complete the task given enough time and resources, the task may be unrealistic or too ambitious. Prove you can get the basics down before activating expert mode.

选择一些你可以教给聪明实习生的东西。如果你最优秀的实习生在给定足够的时间和资源的情况下仍然无法完成任务，那么这个任务可能是不切实际或过于雄心勃勃的。在激活专家模式之前，证明你可以掌握基础知识。

Start by coming up with 5-10 concrete examples of the task. This serves two purposes:

首先想出5-10个具体的任务示例。这有两个目的：

• First, it validates that your idea is well-scoped - not too trivial or vague
• 首先，它验证了您的想法是否范围明确 - 不太琐碎或模糊
• Second, gives you a benchmark for measuring performance later.
• 其次，为后续性能测量提供基准。

Example: Building an Email Agent

示例：构建一个电子邮件代理

At this step, we’d define what tasks our agent needs to handle, which likely includes:

在这一步中，我们将定义我们的代理需要处理的任务，这可能包括：

• Prioritize urgent emails from key stakeholders
• 优先处理来自关键利益相关者的紧急电子邮件
• Schedule meetings based on calendar availability
• 根据日历可用性安排会议
• Ignore spam or emails that don't require responses
• 忽略垃圾邮件或不需要回复的邮件
• Answer product questions based on company documentation
• 根据公司文档回答产品问题

Red flags to avoid:

需要避免的红旗：

• If you can’t come up with concrete examples, your scope is probably too broad.
• 如果你无法提出具体的例子，你的范围可能太广。
• Using an agent when traditional software would work better (e.g., when the logic i...