Cloudflare turned on default AI bot blocking across all its plans this year. Server logs are the only reliable way to tell whether it's affecting you. So we shipped two new integrations that get your Cloudflare logs into Ansehn in minutes, no matter which plan you're on.
Two new integrations went live this week:
- Cloudflare Logpush for customers on Enterprise and Business plans.
- Cloudflare Worker for everyone else on Cloudflare, including the free tier.
Together with the AWS CloudFront native integration and the Custom Logs API we already shipped, that means Ansehn's Server Logs Analytics now works on the infrastructure hosting most B2B websites. If you have been waiting for the connector that fits your stack, this is probably it.
Here is why the timing matters, which integration to pick, and what to look at first once your logs are flowing.
Why this release matters right now
In mid-2026, Cloudflare enabled its "Block AI Bots" WAF setting by default across every plan, including the free tier (source: Cloudflare's own blog). The block happens at Cloudflare's edge, which means it fires before requests reach your origin server. Your robots.txt is bypassed entirely.
If your site is behind Cloudflare and nobody on your team has touched the WAF configuration this year, there is a real chance GPTBot, ChatGPT-User, ClaudeBot, and PerplexityBot are being returned a 403 at the edge, and you have no way to know. Google Analytics does not see AI crawlers. Cloudflare's own dashboards show you blocked bot volume in aggregate, not by page. Only your server logs contain the ground truth, and until now, getting them into Ansehn required Enterprise-level Logpush access or manual uploads.
That gap is closed as of this week.
For background on why AI crawlers matter and which user agents to watch, Kevin's earlier posts cover the ground in more detail:
Which integration to pick
Five paths, ordered by who they fit. Read only the one that matches your setup.
On Cloudflare Enterprise or Business? Use Logpush.
Cloudflare Logpush is Cloudflare's native log-forwarding service. Configure a Logpush job pointed at Ansehn's ingest endpoint, and your request logs start flowing within minutes. No infrastructure change on your side, no scripts. This is the cleanest path if your plan already supports it. Setup guide →
On Cloudflare Free or Pro? Use the Cloudflare Worker.
Logpush is Enterprise-plus only. Rather than push you to upgrade, we built a lightweight Cloudflare Worker that streams request logs to Ansehn from inside your Cloudflare edge. Deploy it once, no plan upgrade required, no downstream infrastructure changes. Setup guide →
On AWS CloudFront? Use the native Firehose integration.
This one has been live for a while. CloudFront streams log events to an Amazon Data Firehose stream you control, and Firehose forwards them to Ansehn over HTTPS. Real-time, zero latency added for your users. Setup guide →
On Fastly, Nginx, Vercel, GCP, Azure, managed WordPress, or anything else? Use the Custom Logs API.
Native integrations for Fastly, Nginx, and Vercel are on our near-term roadmap. Until they ship, the Custom Logs API is the path. You send access logs to our endpoint on whatever schedule fits your stack, we handle the parsing. Setup guide →
Just want to test before committing to a live feed?
Drop a .log file into your Ansehn dashboard for a one-time analysis. Fastest way to see what you have been missing before you touch any infrastructure. Setup guide →
All log forwarding, on every integration, happens off the critical path of your site. Nothing is added to your browser. Nothing touches an end-user request.
What to check first, once your logs are flowing
Three things worth looking at in your first hour with Server Logs Analytics.
First: are the crawlers you expect actually arriving?
Open Top Crawlers. Scan the list. Which AI engines are visible, and which are missing? If ClaudeBot, GPTBot, or PerplexityBot does not appear at all, one of three things is happening:
- Your
robots.txtis blocking the crawler. - Your WAF (Cloudflare, AWS WAF, Akamai, others) is blocking it at the edge, even though
robots.txtallows it. Given the mid-2026 Cloudflare default change, this is now the most common cause. - Your managed host is returning 429 or 403 on some percentage of AI crawler requests before they hit your origin. Search Engine Land documented this pattern on WP Engine earlier this year (source: Search Engine Land, May 2026).
Server Logs Analytics does not diagnose which of the three it is. What it does is confirm the crawler is not arriving, which is the piece your existing stack cannot tell you. Once you know a bot is missing, the fix is technical: check robots.txt, then WAF rules, then open a ticket with your host.
Second: which of your pages is doing the heaviest AI lifting?
Open Top Pages. Sort by ChatGPT-User, Claude-User, or Perplexity-User rather than by their training-bot counterparts. That user-agent traffic is your best proxy for how often real buyers are being served your content inside an AI answer. Pages at the top of that list are the ones AI models keep coming back to. Publish more like them.
Third: what changed this week?
Open Daily Visits. Look for spikes or drops that align with content you shipped, robots.txt edits, or firewall changes. This is where you catch a crawler dropping off before it becomes a citation problem.
The Overview docs walk through each dashboard in more detail.
What is next
Three items already in the near-term roadmap:
- Fastly, Nginx, and Vercel native integrations. These are the biggest gaps in the current portfolio.
- Deeper citation correlation. Server logs show which pages get fetched. Ansehn's Citation Analytics shows which pages get cited in AI answers. The connection between the two is on the roadmap so you can see when a crawl turned into an actual citation.
- Alerts. If a crawler you care about disappears from your logs, or a new one starts hitting your site, you should not have to notice manually.
If there is a use case we have not covered, tell us. The questions you ask are how we decide what to build next.
FAQ
Which Cloudflare integration should I use, Logpush or the Worker?
Logpush if your plan is Enterprise or Business. It uses Cloudflare's native log-forwarding infrastructure and requires no code deployment on your side. The Worker if you are on Free or Pro. It is lightweight and delivers the same data to Ansehn, without needing to upgrade your Cloudflare plan.
Will the Cloudflare Worker slow down my site?
No. The Worker runs at Cloudflare's edge alongside your existing traffic and forwards logs asynchronously. End-user requests are not delayed.
Do I need to disable Cloudflare's "Block AI Bots" setting to use this?
No. The integrations show you what Cloudflare is doing regardless of the setting. If Block AI Bots is enabled, you will see the AI crawlers being blocked (with 403 responses) in your logs. That is the diagnostic value. Whether to keep the setting on is a separate strategic decision.
How long until I see data?
Typically within 60 to 90 seconds of a request hitting your site, for all integrations. The Manual Upload option processes as soon as the file is uploaded.
What if I am on Fastly, Nginx, or Vercel?
Native integrations for those platforms are on the roadmap. Until they ship, use the Custom Logs API. It is more setup than a native integration, but the data you get is the same once it is connected.
Which plans include Server Logs Analytics?
Server Logs Analytics is included on every Ansehn plan. If you are an existing customer on Cloudflare, the two new integration options are already available in your dashboard.
Connect your logs today
If you are an Ansehn customer on Cloudflare, the two new integration options are live in your dashboard right now, under Analytics → Server Logs. Fifteen minutes to setup, real data the same day.
If you are not on Ansehn yet, start a free trial and connect a log source before you finish onboarding. Whether Cloudflare is quietly blocking your AI traffic or your content is genuinely reaching the models, you should be the one who knows.