feat: add middleware to track AI coding agent traffic

[caffeinum]

Summary

Adds Next.js middleware to detect and track AI coding agents (Claude Code, Codex, OpenCode, etc.) visiting the docs.

Uses the 2027-track npm package:

import { withAIAnalytics } from "2027-track/next";
export default withAIAnalytics();

How it works

AI coding agents send Accept: text/markdown header when fetching documentation - browsers never do this. The middleware detects this signal and sends anonymized events to an external analytics API.

Detection signals:

• Accept: text/markdown (primary - browsers never send this)
• axios user-agent → Claude Code
• text/markdown with q= weights → OpenCode
• ChatGPT-User user-agent → Codex

Configuration

No configuration needed. The middleware sends events to an external tracking API - no env vars or secrets required in this repo.

Optional: Set AI_ANALYTICS_ENDPOINT="" to disable tracking entirely.

Data collected

• Host (docs.langfuse.com)
• Path (/docs/tracing)
• Agent type (claude-code, opencode, codex, etc.)
• Country (from Vercel headers)
• Timestamp

Privacy: Events are sent server-side from Vercel Edge. Visitor IP addresses never reach the analytics endpoint.

Dashboard

See collected data: screenshot

Package: https://www.npmjs.com/package/2027-track
Source: https://github.com/caffeinum/ai-docs-analytics

---

Important

Add middleware in middleware.ts to track AI coding agents using 2027-track, with route configuration and dependency update in package.json.

• Middleware:
  • Adds middleware.ts to track AI coding agents using 2027-track package.
  • Detects agents via Accept: text/markdown header and specific user-agents.
  • Configures route matching to exclude static assets and images.
• Dependencies:
  • Adds 2027-track to dependencies in package.json.

^{This description was created by for 6767081. You can customize this summary. It will automatically update as commits are pushed.}

[vercel]

@caffeinum is attempting to deploy a commit to the langfuse Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}