📊 数据处理 社区

分析工具数据查询

Dan Shipper 无需 API 或工程师支持,直接让 Claude 浏览 PostHog 查询按钮点击数据,返回 4,000 次点击

★☆☆ 初级 5 min 2026年1月13日
Dan Shipper
Dan Shipper @danshipper

Co-founder & CEO of Every

来源
📋

使用场景

我想快速查询某个特定按钮在某篇文章中的点击次数,但不想麻烦工程师或自己学习 API。

💬

提示词

Find how many people clicked the "Chat with Claude" button in the agent-native architectures guide from PostHog analytics.

预期结果

Claude 通过 Chrome 浏览 PostHog: - 自动导航到正确的仪表板 - 找到相关的事件追踪数据 - 返回结果:**4,000 次点击** 整个过程无需 API 访问或工程支持。

💡

使用技巧

  • 需要启用 Chrome 连接器并登录 PostHog
  • 适合非技术用户快速获取数据洞察
  • 可以查询任何基于网页的分析工具
📝

原帖内容

· 2026-01-13

Cowork is Claude Code for the rest of us. It's a new way to collaborate with Claude that extends Claude's capabilities beyond coding to any kind of knowledge work. I tested it by asking Claude to find data in PostHog about how many people clicked a specific button. Normally, I'd have to ask an engineer or learn the API myself. Claude just opened a browser, navigated to PostHog, found the right dashboard, and told me: 4,000 clicks. It feels much less like a back-and-forth and much more like leaving messages for a coworker. You can start a task, walk away, and come back to finished work. The future of AI assistants is agentic — they don't just answer questions, they do the work.

Original post image 1