Skip to content

Cloudflare Fake Google Bot rule: off by default, what still blocks

Cloudflare's Fake Google Bot managed rule went from Block to Disabled in February 2026. How to check if your zone turns it back on, and how to add an exception.

By Crawlito · 8 min read

In short

  • Cloudflare's WAF release of 10 February 2026 changed the managed rule Anomaly:Header:User-Agent - Fake Google Bot (ID ending 6aa0bef8) from Block to Disabled in the Cloudflare Managed Ruleset.
  • The companion rule Anomaly:Header:User-Agent - Fake Bing or MSN Bot (ID ending c12cf9c8) followed in the WAF release of 1 July 2026, also moving from Block to Disabled.
  • A disabled-by-default rule still runs on a zone where an override enables it, including a ruleset-level override that turns on every rule in the Cloudflare Managed Ruleset.
  • Cloudflare lists Google Cloud services, Bing Webmaster Tools Site Scan and monitoring tools as common false positives, and recommends a narrow exception placed before the rule that executes the managed ruleset.
  • Bot Fight Mode and WAF custom rules are separate layers that can still stop a request with a spoofed Googlebot user agent after the managed rule is off.
On this page

Cloudflare's managed rule Anomaly:Header:User-Agent - Fake Google Bot is no longer on by default. The WAF release of 10 February 2026 changed its action in the Cloudflare Managed Ruleset from Block to Disabled, and the matching Bing rule followed on 1 July 2026. If a request with a Googlebot user agent is still blocked on your zone, either an override has switched the rule back on, or a different layer is doing the blocking: Bot Fight Mode, a custom rule or another security feature.

This post stays with the two managed rules: what they check, how to find out whether your zone still runs them, the false positives Cloudflare documents, and how to write an exception that does not switch them off for everyone. For Cloudflare's numbered error pages (1010, 1015, 1020), see Cloudflare error 1010 and 1020. For the AI crawler controls in the same dashboard, see the Cloudflare AI crawler settings guide.

What the fake bot managed rules check

Cloudflare's troubleshooting page for fake bot rules says the rules compare the request's User-Agent header against known bot patterns, then verify the source with methods such as reverse DNS lookup or IP validation. If the user agent matches a known bot but the source cannot be verified, the request is flagged as a fake bot.

There are two such rules in the Cloudflare Managed Ruleset (ruleset ID ending 376e9aee). Cloudflare's documentation shows only the last eight characters of each rule ID:

Rule nameRule IDChanged to DisabledSource
Anomaly:Header:User-Agent - Fake Google Bot...6aa0bef810 February 2026WAF release 2026-02-10
Anomaly:Header:User-Agent - Fake Bing or MSN Bot...c12cf9c81 July 2026WAF release 2026-07-01

Both changelog entries use the same wording: Cloudflare is "changing the action for this rule from BLOCK to Disabled". Neither entry gives a reason. Neither entry says anything about zones that had already overridden the rules, so check your own configuration rather than assuming the change reached you.

Two scope notes from Cloudflare's Managed Rules overview:

  • The Cloudflare Managed Ruleset is available on Pro, Business and Enterprise plans, not on Free. Free zones get the Cloudflare Free Managed Ruleset. Cloudflare's documentation does not say whether the free ruleset contains a fake bot check, so if you are on Free, look at Security Events rather than assuming either way.
  • Managed Rules run in the http_request_firewall_managed phase, after custom rules and rate limiting rules. A request that a custom rule blocks never reaches the fake bot rule at all.

Cloudflare does not publish the exact checks, but the method it names is close to what Google tells site owners to do themselves. Google's verification guide says to run a reverse DNS lookup on the IP, confirm the name is in googlebot.com, google.com or googleusercontent.com, then run a forward lookup and confirm it returns the same IP:

bash
host 66.249.66.1
# 1.66.249.66.in-addr.arpa domain name pointer crawl-66-249-66-1.googlebot.com.
host crawl-66-249-66-1.googlebot.com
# crawl-66-249-66-1.googlebot.com has address 66.249.66.1

A request from your laptop with a Googlebot user agent fails that test every time, which is exactly what the rule was built to catch. How to verify real AI crawlers applies the same method to GPTBot, ClaudeBot and the rest.

Does your zone still run the rule?

"Disabled by default" describes Cloudflare's default. Your zone can differ. Cloudflare's Cloudflare Managed Ruleset page describes overrides at three levels, and any of them can turn a disabled rule back on:

Override levelWhat it does to the fake bot rules
Ruleset statusEnabling all rules in the ruleset also enables rules that are disabled by default, and every rule added later
TagSetting the status for a tag enables every rule with that tag; API tag overrides also apply to future rules, and the dashboard asks whether to include them
RuleSetting the rule's own status to enabled runs it with its action, or with the action the override sets

Cloudflare's own note on the ruleset page warns against enabling all rules with an override except during a proof of concept, because it may affect legitimate traffic. A zone that did exactly that during an evaluation and never undid it would still be running both fake bot rules today.

Check in the dashboard

  1. Go to Security rules and filter by Managed rules.
  2. Find the rule with an Execute action that deploys the Cloudflare Managed Ruleset, and open it.
  3. Look at the ruleset-level settings. A ruleset status override that enables all rules turns the fake bot rules on.
  4. Select Browse rules, search for Fake Google Bot and Fake Bing, and read the Status toggle for each.

Check with the API

The zone's http_request_firewall_managed entry point ruleset holds the execute rule and any overrides. Cloudflare's exception API page uses this call:

bash
curl "https://api.cloudflare.com/client/v4/zones/$ZONE_ID/rulesets/phases/http_request_firewall_managed/entrypoint" \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN"

Find the rule whose action is execute and whose action_parameters.id is the Cloudflare Managed Ruleset (efb7b8c949ac4650a09736fc376e9aee). Its overrides object is what matters. Per Cloudflare's override documentation, the shape is:

json
{
  "overrides": {
    "enabled": true,
    "categories": [{ "category": "<TAG_NAME>", "enabled": true }],
    "rules": [{ "id": "<RULE_ID>", "enabled": true, "action": "block" }]
  }
}

A top-level "enabled": true, a category override with "enabled": true on a tag the rule carries, or a rule entry whose id ends in 6aa0bef8 or c12cf9c8 with "enabled": true all mean the rule runs. Rule overrides take precedence over tag overrides, which take precedence over ruleset overrides.

How to see which rule blocked the request

Before changing anything, confirm the block came from this rule. In Security > Analytics, open the Events tab, which Cloudflare's Security Events page describes. Filter by the user agent or IP of the blocked request, then check the Service field of the event:

  • Managed rules, with the event pointing to Anomaly:Header:User-Agent - Fake Google Bot or Fake Bing or MSN Bot (rule ID ending 6aa0bef8 or c12cf9c8): this post applies.
  • Bot Fight Mode, Custom rules or anything else: the managed rule is not the cause, and an exception will not help. See the next section.

Remember the limits Cloudflare documents: Free plans see sampled logs only, and events are kept for up to the last 31 days. If the block is older than that, reproduce it and look again.

Documented false positives

Cloudflare's troubleshooting page names three groups of legitimate traffic that these rules can catch:

TrafficWhy it is flagged
Google Cloud services such as Cloud Workflows or Cloud FunctionsThey can send a Google-related user agent from IPs outside the Googlebot range, so IP verification fails
Bing Webmaster Tools Site ScanSite Scan does not use Bingbot's IP range, and Bing's own Verify Bingbot tool does not recognise its addresses
Third-party uptime monitors and testing toolsThey set a bot-like user agent from their own infrastructure

Cloudflare does not say which Google services trigger the rule beyond those examples, but Google's documentation shows why such traffic exists: Google says IPs in its user-triggered-fetchers.json list resolve to gae.googleusercontent.com hostnames and are used, for example, when a site on Google Cloud fetches external RSS feeds for its users. Those requests come from real Google infrastructure but are not Googlebot, so a check that only accepts Googlebot's addresses can reject them.

For Site Scan, Cloudflare's separate page calls it a known issue of Bing Webmaster Tools and recommends a temporary exception, deleted once the scan finishes. That page was last updated in May 2026, before the July change; with the Bing rule now disabled by default, it only applies if your zone enables the rule.

How to add a managed rule exception

If you want the fake bot rules on and need to let one legitimate service through, Cloudflare recommends an exception that skips only the triggered rule for only the affected requests. Turning the rule off for the whole zone is the broad version of the same fix.

In the dashboard, per Cloudflare's exception steps:

  1. Go to Security rules and select Create > Managed rules.
  2. Name the exception and write the expression in When incoming requests match.
  3. Under Then, choose Skip specific rules from a Managed Ruleset, select Select ruleset, then Select rules next to the Cloudflare Managed Ruleset.
  4. Search for Fake Google Bot (or Fake Bing), tick it and select Next.
  5. Select Deploy.

Cloudflare suggests scoping the expression by source IP or range, URI path, or ASN. For a monitoring service with published addresses checking one health endpoint:

(ip.src in {203.0.113.10 203.0.113.11} and http.request.uri.path eq "/health")

Three rules from Cloudflare's documentation decide whether the exception works:

  • Order. A skip rule only applies to execute rules listed after it. If the exception sits below the rule that executes the Cloudflare Managed Ruleset, nothing is skipped.
  • Scope. A zone-level exception does not skip account-level managed rules, and account-level rulesets run first.
  • Product. Exceptions apply to WAF managed rulesets only. To skip Browser Integrity Check or similar features, you need a custom rule with the Skip action instead.

Exceptions also take priority over overrides, so a narrow exception works even on a zone that enables every rule.

Other layers that still stop a spoofed user agent

With the managed rule disabled, a spoofed Googlebot request can still be stopped by:

  • Bot Fight Mode. Cloudflare's Bot Fight Mode page says it identifies traffic matching patterns of known bots and challenges it. It does not run on the Ruleset Engine, so Skip, Bypass and Allow actions in WAF custom rules or Page Rules have no effect on it. The one documented way around it is an IP Access rule that matches the request first.
  • Super Bot Fight Mode on paid plans, which runs after Managed Rules and can be skipped with a custom rule.
  • Custom rules of your own, for example one that matches Googlebot in the user agent and not cf.client.bot.

The test itself is usually the problem. A request from your machine is not Googlebot, whatever header it sends, so a block on your test says little about what the real crawler gets. JA3 and JA4 fingerprints: how bot products identify clients covers what a self-test can and cannot prove, and see your page as an AI crawler with curl shows how to read what comes back.

Frequently asked questions

Is the Cloudflare Fake Google Bot rule still on by default?

No. Cloudflare's WAF release of 10 February 2026 changed it from Block to Disabled in the Cloudflare Managed Ruleset. It still runs on any zone where an override enables it.

Did the change also cover the Fake Bing or MSN Bot rule?

Yes, separately. The WAF release of 1 July 2026 changed Anomaly:Header:User-Agent - Fake Bing or MSN Bot from Block to Disabled.

Will a managed rule exception fix a block from Bot Fight Mode?

No. Exceptions apply only to WAF managed rulesets, and Bot Fight Mode cannot be skipped with WAF custom rules or Page Rules either. Check the Service field in Security Events to see which feature acted.

Does the fake bot rule block the real Googlebot?

It is designed not to: it verifies the source with methods such as reverse DNS or IP validation and flags only requests it cannot verify. Cloudflare's documented false positives are other Google services, Bing Site Scan and monitoring tools, not Googlebot itself.

Check what crawlers receive from your zone

After you change an override or add an exception, test from outside your own network. The free check tests a single page with no account needed. If several zones or rule sets are involved, an audit covers up to 500 pages and comes with a PDF of evidence and prioritized fixes.

Filed under CDN and bot protection. Topics: cloudflare, waf, bot-protection, googlebot, troubleshooting