Cloudflare Docs
Learning Paths
Secure your Internet traffic and SaaS apps (Learning Path)
Edit this page on GitHub
Set theme to dark (⇧+D)

Recommended HTTP policies

  3 min read

We recommend you add the following HTTP policies to build an Internet and SaaS app security strategy for your organization.


Bypass HTTP inspection for applications that use embedded certificates. This will help avoid any certificate pinning errors that may arise from an initial rollout.

ApplicationinDo Not InspectDo Not Inspect

Bypass HTTPS inspection for Android applications (such as Google Drive) that use certificate pinning, which is incompatible with Gateway inspection.

ApplicationinGoogle DriveAndDo Not Inspect
Passed Device Posture ChecksinOS Version Android (OS version)

Bypass HTTP inspection for a custom list of domains identified as incompatible with TLS inspection.

Domainin listDomainInspectionBypassOrDo Not Inspect
Domainin listKnown Domains

Block security categories, such as Command and Control & Botnet and Malware, based on Cloudflare’s threat intelligence.

Security RisksinAll security risksBlock

Entries in the security risk content subcategory, such as New Domains, do not always pose a security threat. We recommend you first create an Allow policy to track policy matching and identify any false positives. You can add false positives to your Trusted Domains list used in All-HTTP-Domain-Allowlist.

After your test is complete, we recommend you change the action to Block to minimize risk to your organization.

Content CategoriesinQuestionable Content, Security Risks, Miscellaneous, Adult Themes, GamblingAllow

Block specific domains or hosts that are malicious or pose a threat to your organization. Like All-HTTP-ResolvedIP-Blocklist, this blocklist can be updated manually or via API automation.

Domainin listDomain BlocklistOrBlock
Hostin listHost BlocklistOr
Hostmatches regex.*example\.com

Block unauthorized applications to limit your users’ access to certain web-based tools and minimize the risk of shadow IT. For example, the following policy blocks popular AI chatbots.

ApplicationinChatGPT, BardBlock

Isolate traffic for privileged users who regularly access critical systems or execute actions such as threat analysis and malware testing.

Security teams often need to perform threat analysis or malware testing that could trigger malware detection. Likewise, privileged users could be the target of attackers trying to gain access to critical systems.

User Group NamesinPrivileged UsersIsolate

Restrict access for users included in an identity provider (IdP) user group for risky users. This policy ensures your security team can restrict traffic for users of whom malicious or suspicious activity was detected.

Destination IPnot in listQuarantined-Users-IPAllowlistAndBlock
User Group NamesinQuarantined Users

Isolate high risk domains or create a custom list of known risky domains to avoid data exfiltration or malware infection. Ideally, your incident response teams can update the blocklist with an API automation to provide real-time threat protection.

Content CategoriesinNew Domain, Newly Seen DomainsOrIsolate
Domainin listDomain Isolation