Reading the Bots screen
Bots shows what crawlers did on your site: which ones came, what they fetched, and whether they came to answer someone or to train a model. It appears in a site’s navigation once your server has reported a crawler hit; a site with none carries no empty tab.
Every number here is a request to your server, counted by UTC day. Filters, comparison and the site’s timezone do not apply: a crawler hit has no country, device or referrer, and a day is a day everywhere.
Families
The first line is the pool of crawler hits in the range, split by what the crawler was doing. The split between the first two is the point of the screen.
| Family | What it means | Examples |
|---|---|---|
| Answers | Fetching a page right now, to answer a person’s question | ChatGPT-User, OAI-SearchBot, PerplexityBot, Claude-User |
| Training | Collecting pages for a model | GPTBot, ClaudeBot, CCBot, Bytespider |
| Search | Classic indexing | Googlebot, bingbot, DuckDuckBot, Applebot |
| Social | Unfurling a link someone pasted in a chat or a feed | Slackbot, Discordbot, Twitterbot, facebookexternalhit |
| SEO | Third-party crawlers that sell what they find | AhrefsBot, SemrushBot |
| Other | Recognised as a bot, but not on the list by name | monitors, HTTP libraries, anything new |
The sentence under the bar says which side won, names the busiest crawler, and says how many crawlers read your llms.txt, if any did.
Crawlers
Ranked by hits, with the family beside each name, the number of distinct pages it touched, and when it was last seen. A crawler that walked the whole site shows many pages; one that fetches the same three documentation pages on demand shows few.
A crawler is what its user agent claimed. Lynq does not verify that a request calling itself Googlebot came from Google, so a scraper wearing a famous name is counted under that name.
Looking for instructions
robots.txt, llms.txt and sitemap fetches are listed separately from pages, because a crawler asking for them is asking what it may do. A crawler that reads robots.txt and then fetches pages it was told not to is visible by comparing the two tables.
Pages
Ranked by crawler hits, with how many different crawlers fetched each. The Answers, Training and Search views narrow the table to that family, so you can see which pages ChatGPT and Perplexity fetch to answer people, as opposed to what Googlebot indexes. The view is in the address bar, so a link carries it.
What Lynq does not do
It does not block or slow any crawler; that is a firewall’s job, and robots.txt is where you state your wishes. It does not count people: a request from a browser is never sent by the middleware, so visitor numbers on the other screens are unaffected, and a visitor who opted out stays out here too.