AI crawlers, search bots and training agents
The 63 crawlers theBizness.ai recognizes: who runs them, why they visit, and how to see when they hit your site.
AI answers
Fetches made when an AI assistant opens a page to answer someone, right now.
| Crawler | Operator |
|---|---|
ChatGPT-User | OpenAI |
Claude-User | Anthropic |
Perplexity-User | Perplexity |
Google-Agent | |
Google-NotebookLM | |
Google-Read-Aloud | |
Copilot | Microsoft |
MistralAI-User | Mistral |
Amzn-User | Amazon |
DuckAssistBot | DuckDuckGo |
xAI-SearchBot | xAI |
Grok-DeepSearch | xAI |
meta-externalfetcher | Meta |
Kimi-User | Moonshot AI |
Qwen-User | Alibaba |
Search indexes
Crawlers that discover and index pages for search engines and AI search.
| Crawler | Operator |
|---|---|
OAI-SearchBot | OpenAI |
Claude-SearchBot | Anthropic |
PerplexityBot | Perplexity |
Googlebot | |
Google-InspectionTool | |
Bingbot | Microsoft |
msnbot | Microsoft |
MistralAI-Index | Mistral |
Amzn-SearchBot | Amazon |
meta-webindexer | Meta |
Kimi-SearchBot | Moonshot AI |
TikTokSpider | ByteDance |
Baiduspider | Baidu |
YouBot | You.com |
DuckDuckBot | DuckDuckGo |
YandexBot | Yandex |
Training crawlers
Crawlers that collect public pages that may be used to train AI models.
| Crawler | Operator |
|---|---|
GPTBot | OpenAI |
ClaudeBot | Anthropic |
GoogleOther | |
Google-CloudVertexBot | |
Google-Extendedrobots.txt token | |
Applebot | Apple |
Applebot-Extendedrobots.txt token | Apple |
Amazonbot | Amazon |
meta-externalagent | Meta |
KimiBot | Moonshot AI |
Bytespider | ByteDance |
ERNIEBot | Baidu |
QwenBot | Alibaba |
DeepSeekBot | DeepSeek |
ChatGLM-Spider | Zhipu AI |
cohere-ai | Cohere |
cohere-training-data-crawler | Cohere |
AI2Bot | Allen AI |
CCBot | Common Crawl |
Other AI bots
Ad checks, link previews and other automated fetchers run by AI companies.
| Crawler | Operator |
|---|---|
OAI-AdsBot | OpenAI |
GrokBot | xAI |
xAI-Bot | xAI |
xAI-Grok | xAI |
xAI-Web-Crawler | xAI |
Grok | xAI |
meta-externalads | Meta |
facebookexternalhit | Meta |
FacebookBot | Meta |
Doubaobot | ByteDance |
YiyanBot | Baidu |
TongyiBot | Alibaba |
AliyunBot | Alibaba |
See crawlers in your analytics
Crawlers don't run JavaScript, so report them from your server. Copy your ingest secret and website ID from Dashboard → Setup into environment variables, then add this middleware. Crawler visits appear in the Crawlers card on your dashboard, by crawler and by page. Requests from regular visitors are ignored, and crawls never count toward your event limit.
Next.js
// middleware.ts
import { NextResponse, type NextRequest, type NextFetchEvent } from 'next/server';
const MAYBE_BOT = /bot|crawl|spider|-user|agent|fetcher|copilot|grok|notebooklm|read-aloud|externalhit|cohere/i;
export function middleware(request: NextRequest, event: NextFetchEvent) {
const userAgent = request.headers.get('user-agent') ?? '';
if (MAYBE_BOT.test(userAgent)) {
event.waitUntil(
fetch('https://thebizness.ai/api/crawls', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
Authorization: `Bearer ${process.env.THEBIZNESS_INGEST_SECRET}`,
},
body: JSON.stringify({
websiteId: process.env.THEBIZNESS_WEBSITE_ID,
url: request.nextUrl.pathname,
userAgent,
}),
}).catch(() => {})
);
}
return NextResponse.next();
}
export const config = { matcher: ['/((?!_next/|favicon.ico).*)'] };Any server
Send websiteId, url and the request's userAgent to POST /api/crawls with your ingest secret. Add the response status if you have it, to see which pages crawlers hit that return 404.
curl -X POST https://thebizness.ai/api/crawls \
-H "Authorization: Bearer $THEBIZNESS_INGEST_SECRET" \
-H "Content-Type: application/json" \
-d '{"websiteId":"YOUR_WEBSITE_ID","url":"/pricing","userAgent":"Mozilla/5.0 (compatible; GPTBot/1.2)","status":200}'Crawler FAQ
Why doesn't my analytics show crawlers already?
Crawlers fetch your HTML without running JavaScript, so no browser-based analytics script ever sees them. You have to report them from your server, which is what the snippet below does.
Do crawler visits count toward my plan's event limit?
No. Crawls are stored separately from pageviews and do not count toward your event limit or your visitor numbers.
Can a crawler's user agent be faked?
Yes. Anyone can send a user agent that says GPTBot. We identify crawlers by their user agent, which is what most of them announce, but we do not verify the request came from the operator's IP ranges.
What is the difference between Google-Extended and Googlebot?
Googlebot crawls pages for Google Search. Google-Extended is only a robots.txt token that controls whether Google may use your content for Gemini; it never appears in a user agent. Applebot-Extended works the same way for Apple.
Know when AI reads your site
- See which assistants fetch your pages to answer people
- Spot training crawlers and the pages they take
- Find the pages bots hit that return 404
14-day free trial. No card required.