Live metrics
The corpus
| Reference resources | 85 |
|---|---|
| Editorial guides | 69 (118,237 words) |
| Topic clusters complete | 5 of 6 |
AI-crawler volume (30-day rolling)
| Date (UTC) | AI total | GPTBot | OAI-SearchBot | ChatGPT-User | ClaudeBot | Claude-User | PerplexityBot | Google-Extended | Bytespider | All worker hits |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-09-05 | 1743 | 148 | 37 | 1065 | 112 | 25 | 356 | 0 | 0 | 59321 |
| 2026-08-29 | 1699 | 153 | 40 | 1097 | 115 | 25 | 269 | 0 | 0 | 56146 |
| 2026-08-22 | 1826 | 291 | 22 | 1142 | 114 | 27 | 230 | 0 | 0 | 52830 |
| 2026-08-15 | 1818 | 358 | 22 | 1056 | 116 | 27 | 239 | 0 | 0 | 55152 |
| 2026-08-13 | 1837 | 359 | 25 | 1057 | 118 | 29 | 249 | 0 | 0 | 55660 |
Reports
The full per-bot and per-page breakdown behind the table above is published monthly at the AI-crawler traffic report (Markdown: .md, JSON: .json).
How this is measured
The worker logs every request to Cloudflare's Analytics Engine; a scheduled GitHub Actions job queries it weekly and commits the snapshot this page renders. Asset-served paths (/llms.txt, robots.txt, sitemaps) bypass the worker and are deliberately uncounted, so treat these numbers as a floor. Want this for your own site? Start with measuring AI agent traffic.