[invisit](https://invisit.dev) · [Docs](https://invisit.dev/docs) · [llms.txt](https://invisit.dev/llms.txt) · [llms-full.txt](https://invisit.dev/llms-full.txt)

# User agent lookup

> Paste a user agent from your logs and see what sent it: a person’s browser, a named crawler and who runs it, or automation, and whether invisit would count it as a visit.

- Run it: https://invisit.dev/tools/user-agent-lookup (free, no account)
- Last changed: 2026-11-05

## How the lookup decides

It runs three sets of rules, in this order, the same ones invisit uses to sort traffic:

1. **Known crawlers.** A list of about 80 bots by name, each with the company that runs it and what it is for: AI training, AI search, AI assistants, search engines, SEO tools, link previews and archivers.
2. **Automation.** Headless browsers, test runners, uptime monitors and HTTP libraries such as curl or Python’s requests. These are kept out of visits even when they don’t say they’re a bot.
3. **People.** Anything left is read as a browser, a system and a kind of device.

A match in the first list doesn’t prove the bot is real, only that the request says so. The bot pages explain how to check.

## Where to find user agents

Your web server’s access log has one per request. In Nginx and Apache’s default combined format it is the last quoted field on each line. On Vercel, Netlify and Cloudflare, the request logs show it as `user-agent`. In your own browser, the button above fills in yours.

## Common questions

### Can a user agent be faked?

Yes, trivially. A user agent is a header the client writes itself, so a scraper can claim to be Googlebot. To know a crawler is real, check the request’s IP address against the ranges or the reverse DNS names its operator publishes. Each bot page here says how.

### Is anything I paste stored or sent anywhere?

No. The lookup runs in your browser with the same rules invisit’s collector uses, and nothing you paste leaves the page.

### Why does it say invisit would not count a crawler as a visit?

invisit keeps bots out of your visitor numbers. Crawler reads can be recorded separately, through the server SDK or a Cloudflare Worker, and show up in their own report.

### What is the difference between ChatGPT-User and GPTBot?

GPTBot crawls pages that may be used to train OpenAI’s models. ChatGPT-User fetches a page because someone asked ChatGPT about it right then. They are separate bots with separate robots.txt tokens, so you can block one and allow the other.