AI crawlers

What you see

• visits per crawler per day, for 7, 30 or 90 days
• each crawler with its company, purpose (training, search or live fetch), last visit and top pages
• your sitemap pages no AI crawler read in 30 days
• crawlers your robots.txt turns away

Echo knows the crawlers of OpenAI, Anthropic, Perplexity, Google, Apple, Meta, Amazon, Mistral, DuckDuckGo, Common Crawl and ByteDance.

Set it up

Go to Website → Health → AI crawlers → Setup. Choose one way:

A snippet on your server (required: one of these ways)

1. Click Make a key. Copy it now: it's shown once.
2. Pick your host: Cloudflare Worker, Next.js, Vercel, Netlify, Node or PHP.
3. Copy the snippet (it already has your key) and add it to your site.
4. Deploy. The page says when the first visit arrives.

Or send your logs

• Cloudflare Logpush: free on every Cloudflare plan. Echo gives you the address, the fields and a filter that keeps only AI agents.
• Vercel log drains: on Vercel Pro or Enterprise.
• A custom endpoint: any host or log shipper, as JSON lines, CSV or access logs.
• Upload a log file for a one-off look.

Sites on Framer can't run a snippet. Use the AI score and pages AI can read instead, or put Cloudflare in front of the site.

What Echo stores

The crawler, the page's path (never the query string), the status and the time. Nothing about human visitors. Single visits are kept 90 days.

Actions

A robots.txt that blocks AI search crawlers becomes an action, checked daily, so a fix closes it the next day. Key pages (pricing, comparisons, product, FAQ) no AI search crawler read in 30 days get an action too.