展示HN：我扔掉了我的分析仪表板，用42个MCP工具替代了它
在十月份，我开发了一个网络分析工具。它有两个显著特点：价格便宜，并且提供每周的人工智能邮件摘要，这样你就不必频繁查看仪表板。结果发现，价格便宜并没有太大影响，但我那几个用户非常喜欢这个邮件摘要。
如今，世界变化迅速，我突然发现自己开始尝试使用代理，并在Claude Code中完成所有工作，因此我决定抛弃一切，重新构建这个工具，命名为lodd.dev，这是一个为代理设计的无头网络分析工具。它包含42个MCP工具和一个完整的API，允许代理在其流程中调用和处理分析数据。
我尽量优化它以适应代理的使用，因此响应效率很高（快照大约60个tokens），还提供了一个llms.txt文件来指导代理进行设置，简化身份验证的只读模式，以及一个用于桌面和移动应用的托管OAuth版本和一个用于终端使用的标准输入版本，后者使用API密钥进行身份验证。人机交互的身份验证使用邮件中的一次性密码（OTP）。
明确要求代理使用这个工具效果很好，我特别喜欢将其与其他上下文来源（如GSC或提交历史）结合使用，但让代理在没有提示的情况下调用它有点挑战。不过，在Claude.md文件顶部添加几行内容确实有所帮助。
我很想听听你们对能够在同一对话中获取分析数据（或其他自动化工作流程）有什么价值的看法。
查看原文
Back in October, I built a web analytics tool. It had two distinguishing features: it was cheap, and it had a weekly AI email summary so you wouldn't have to check your dashboards. Turned out that cheap made no difference, but the few users I had really liked the email digest.<p>The world moves fast these days, and suddenly I found myself experimenting with agents, and doing all my work in Claude Code, so I decided to throw out everything and rebuild the tool as lodd.dev, a headless web analytics tool for agents. 42 MCP tools and a full API that lets agents call and act on analytics data as part of their processes.<p>I've tried to optimise it for agent usage, so efficient responses (a snapshot is 60ish tokens), an llms.txt to guide the agent to handle setup, read only for simplified auth, and both a hosted oAuth (for desktop and mobile apps) and stdio version using an API key for terminal use. Human in the loop authentication using OTP in an email.<p>Explicitly asking the agent to use it works really well, and I particularly like combining with other context sources like GSC or commit history, but it's been a little bit of a challenge having the agent call it unprompted. A few lines at the top of a Claude.md file has made a difference though.<p>I'd love to hear what value, if any, you see from being able to get analytics data in the same conversation as your code (or other automatic workflows)?