AI Crawler Monitoring When one line locks out AI bots, searchVIU notices.

searchVIU checks your robots.txt every 15 minutes for search engines and AI bots. If a change locks out a bot, you see the affected URLs and their Google clicks.

Every 15 minutesevery robots.txt of your hostsBot classesSearch, AI search, AI training, fetches on user requestGoogle clicks on these URLsfrom Search Console
robots.txt · bot matrixexample-shop.com · checked 10:15
GooglebotSearch
allowed
OAI-SearchBotAI search
allowed
Claude-SearchBotAI search
allowed
PerplexityBotAI search
allowed
ChatGPT-UserFetch on user request
allowed
GPTBotAI training · own group
blocked
Why AI bots visit

Every AI bot has its purpose, from training and search to answers.

OpenAI sends GPTBot for training, OAI-SearchBot for search and ChatGPT-User when someone asks a question in ChatGPT. Each bot has its own setting. A rule for all bots, however, also hits the bots that fetch for search.

One line too manyA new user agent slips into the group with Disallow: / and is locked out.
Stricter bot protectionThe firewall blocks bots that the robots.txt allows. Only the logs reveal this.
robots.txt goes downIf the file responds with a server error, bots quickly treat it like a block.
Bot matrix

Which bot is allowed in, and what does a block hit?

searchVIU assigns every known bot to a class, from search, AI search, AI training and fetches on user request to SEO tools. For each bot you see whether it is allowed in, whether it has its own group and how many URLs a block hits. That way you can block training and let AI search in.

BotClassAccessBlocked URLsGoogle clicks on these URLs
GooglebotSearchallowed00
OAI-SearchBotAI searchblockedis in the GPTBot group48,21031,400
Claude-SearchBotAI searchallowed00
PerplexityBotAI searchallowed00
ChatGPT-UserFetch on user requestallowedaccording to OpenAI, does not always follow the robots.txt00
GPTBotAI trainingblockedintended48,21031,400

Demo project. The clicks from Search Console show how important the blocked pages are.

More about robots.txt monitoring
Every 15 minutes

searchVIU saves every version and compares it line by line.

searchVIU compares every new robots.txt line by line with the previous one, per user agent. If a change blocks a search bot from Google or an AI search, the alert is critical. For training bots there is a warning, and for fetches on user request a notice with the severity Info. Each check run sends at most one email per project, and known changes are not emailed again.

Go to the alert list
robots.txtexample-shop.com · change found
+ User-agent: OAI-SearchBotUser-agent: GPTBot
Disallow: /

User-agent: *
Disallow: /basket/
Alert · CriticalOAI-SearchBot blocked · 48,210 URLs with 31,400 Google clicks
Day to day

You test a new robots.txt before rolling it out.

For teams with many hosts and frequent deployments.

Test modeYou can run through a planned robots.txt beforehand. Nothing is saved, and no email goes out.
All hostsEvery host of your live crawls is checked, from the subdomain to the old domain after a migration.
Content SignalsWith these entries in the robots.txt, you set whether content may be used for search, AI answers or training. If they change, searchVIU reports that too.
robots.txt and logs

The logs show whether an allowed bot really visits.

Your server logs show whether ChatGPT-User or ClaudeBot really fetch your pages, how often and which ones. They also reveal when the firewall blocks bots that the robots.txt allows.

Go to AI log file analysis
Rendering

What is left without JavaScript?

A bot only sees content that JavaScript adds to the page if it renders the page. searchVIU crawls every page in a real Chrome and compares the source code with the rendered page. That way you see what a bot finds without rendering.

    More about JavaScript SEO
    /hiking-boots/example-shop.com · example
    On the pageSource codeRendered
    Products036
    Internal links18142
    Body text25 words410 words
    A quiet day

    The right bots are allowed in.

    • AI search allowedOAI-SearchBot, Claude-SearchBot, PerplexityBot
    • Training under controlGPTBot blocked in its own group
    • robots.txt unchanged96 checks since yesterday
    Today · all quiet

    Frequently asked questions about AI crawler monitoring

    Does searchVIU have its own AI crawler?

    No. AI crawler monitoring means that searchVIU monitors which AI bots may fetch your website. The searchVIU crawler is called searchVIUbot and crawls your website for SEO Monitoring.

    Which bots does searchVIU recognize?

    searchVIU knows the common bots for search, AI search, AI training, fetches on user request and SEO tools. These include, for example, Googlebot, GPTBot, OAI-SearchBot, Claude-SearchBot, PerplexityBot, DuckAssistBot, ChatGPT-User, Claude-User and MistralAI-User. According to their operators, ChatGPT-User, Perplexity-User and meta-externalfetcher do not always follow the robots.txt. searchVIU therefore reports a block for these bots as a notice. According to Anthropic, Claude-User can be blocked via robots.txt. Go to the AI bot list

    We have many domains and subdomains. Does that work?

    Yes. searchVIU checks the robots.txt of every host from your live crawls, including country sites and old domains after a migration.

    How quickly does the notification arrive?

    searchVIU checks every 15 minutes. If a check run finds a change, you get at most one notification per project, like all alerts by email, Slack, Teams or webhook. searchVIU does not report known changes again.

    Is this included in the price?

    Yes. AI crawler monitoring is part of SEO Monitoring, from €399 per month with 50,000 URLs included. You can find all prices on the pricing page. Go to pricing

    How is this different from AI log file analysis?

    AI crawler monitoring shows which bots may fetch your pages. AI log file analysis shows which bots actually visit and how often. It is an add-on to SEO Monitoring. Go to AI log file analysis

    Demo

    Let's talk about your website.

    In the demo, Michael shows you how searchVIU works. We use your website and a demo project for this.

    1. Request in three short steps
    2. Quick alignment on crawler access, NDA and Search Console
    3. Demo with your website and the demo project
    4. Start planned and supported together
    1. Website
    2. Needs
    3. Contact

    Which website is it about?