导航|地址|刷新
Get started with AI Security for Apps
图
Skip to contentDocumentation Index
Fetch the complete documentation index at: https://developers.cloudflare.com/waf/llms.txt
Use this file to discover all available pages before exploring further.



DocsDirectoryAPISDKsChangelog

Search

Log in






WAF

/




Overview

Get started

Concepts


Traffic detections


Overview

Attack score


Leaked credentials


Overview

Get started

Common API calls

Terraform examples

Example mitigation rules






Malicious uploads


Overview

Get started

Example rules

Common API calls

Terraform examples






AI Security for Apps


Overview

Get started

PII detection

Unsafe and custom topic detection

Prompt injection detection

Token counting

Example mitigation rules

Log mode vs production mode

Available fields





Bot score ↗


Threat intelligence


Overview

Get started

Example rules

Available fields










Custom rules


Overview

Create in the dashboard

Create via API

Create using Terraform ↗


Configure a rule with the Skip action


Overview

API examples

Skip options






Common use cases


Allow traffic from IP addresses in allowlist only

Allow traffic from search engine bots

Allow traffic from specific countries only

Block Microsoft Exchange Autodiscover requests

Block requests by attack score

Block traffic by geographical location

Block traffic from specific countries

Build a sequence rule within custom rules

Challenge bad bots

Configure token authentication

Exempt partners from Hotlink Protection

Issue challenge for admin user in JWT claim based on attack score

Require a specific cookie

Require known IP addresses in site admin area

Require specific HTTP headers

Require specific HTTP ports

Stop R-U-Dead-Yet? (R.U.D.Y.) attacks

Update custom rules for customers or partners





Custom rulesets






Rate limiting rules


Overview

Request rate calculation

Create in the dashboard

Create via API

Create using Terraform ↗

Find appropriate rate limit

Rate limiting parameters

Rule examples

Best practices

Troubleshooting






Managed rules


Overview

Deploy in the dashboard

Deploy via API

Deploy using Terraform ↗

Troubleshooting


Create exceptions


Overview

Add an exception in the dashboard

Add an exception via API






Log the payload of matched rules


Overview

Configure in the dashboard

View the payload content in the dashboard

Configure via API

Store decrypted matched payloads in logs


Command-line operations


Overview

Generate a key pair

Decrypt the payload content










Check for exposed credentialsDeprecated


Overview

How it works

Configure via API

Configure using Terraform

Test your configuration

Monitor exposed credentials events

Upgrade to leaked credentials detection






Rulesets reference


Cloudflare Managed Ruleset


Cloudflare OWASP Core Ruleset


Overview

Concepts

Evaluation example

Configure in the dashboard

Configure via API

Configure in Terraform ↗





Cloudflare Exposed Credentials Check Managed RulesetDeprecated

Cloudflare Sensitive Data Detection










Additional tools



Lists


Overview

Custom lists

Bulk Redirect Lists ↗

Managed Lists

Create in the dashboard

Use lists in expressions


Lists API


Overview

JSON object

Endpoints










IP Access rules


Overview

Create a rule

Parameters

Actions






Scrape Shield


Overview

Email Address Obfuscation

Hotlink Protection





User Agent Blocking

Zone Lockdown

Browser Integrity Check

Enable security.txt ↗

Privacy Pass

Replace insecure JS libraries

Security Level

Validation checks






Account-level configuration


Overview


Custom rulesets


Overview

Use the dashboard

Use the API

Use Terraform ↗






Rate limiting rulesets


Overview

Create in the dashboard

Create via API

Create using Terraform ↗






Managed rulesets


Overview

Deploy in the dashboard

Deploy via API

Deploy using Terraform ↗

Create exceptions ↗










Analytics


Security Analytics

Security Events





Security features interoperability


Reference


Alerts

Phases


Legacy features



WAF managed rules (previous version)


Overview

Troubleshooting

WAF managed rules upgrade






Rate Limiting (previous version)


Overview

Troubleshooting

Rate limiting upgrade





Firewall rules ↗

Firewall rules upgrade










Troubleshooting


Bing's Site Scan blocked by a managed rule

Fake bot detection blocking legitimate requests

Issues sharing to Facebook

SameSite cookie interaction with Cloudflare

Rule phase interactions

FAQ





Glossary


Changelog


Overview

Changelog

Scheduled changes

Historical (2024)

Historical (2023)

Historical (2022)






Agent resources


Agent setup ↗

Cloudflare Skills ↗

Code Mode MCP Server ↗

Domain-specific MCP Servers ↗MCP

WAF llms.txt ↗

WAF llms-full.txt ↗

Cloudflare Docs llms.txt ↗

Cloudflare Docs llms-full.txt ↗













Home

/WAF

/…
Traffic detections


/AI Security for Apps

/Get started



Get started with AI Security for Apps


Last updated Apr 21, 2026|Copy as Markdown|View as Markdown|Agent setup

Overview1. Turn on AI Security for Apps2. Save or add an LLM-related endpoint3. Add cf-llm label to endpoint4. (Optional) Generate API traffic5. Review labeled traffic and detection behavior6. Mitigate harmful requests





1. Turn on AI Security for Apps








Note


AI Security for Apps (formerly Firewall for AI) is only available in the new application security dashboard.





In the Cloudflare dashboard, go to the Security Settings page.
Go to Settings ↗


(Optional) Filter by Detection tools.



Turn on AI Security for Apps.





Enable the feature using a PUT request similar to the following:
curl "https://api.cloudflare.com/client/v4/zones/$ZONE_ID/ai-security/settings" \--request PUT \--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \--json '{ "enabled": true }'




2. Save or add an LLM-related endpoint


Once you have onboarded your domain to Cloudflare and some API traffic has already been proxied by Cloudflare, the Cloudflare dashboard will start showing discovered endpoints.

Save the relevant endpoint receiving LLM-related traffic to Endpoint Management once it has been discovered, or add the endpoint manually.








In the Cloudflare dashboard, go to the Web assets page.
Go to Web assets ↗


Go to the Discovery tab.



Find the endpoint receiving requests with LLM prompts in the list and select Save next to the endpoint.






Log in to the Cloudflare dashboard ↗, and select your account and domain.

Go to Security > API Shield.

Go to the Discovery tab.

Find the endpoint receiving requests with LLM prompts in the list and select Save next to the endpoint.





If you did not find the endpoint in the Discovery tab, you can add it manually:







Go to the Endpoints tab.

Select Add endpoints > Manually add.

Choose the method from the dropdown menu and add the path and hostname for the endpoint.

Select Add endpoints.





Go to the Endpoint Management tab.

Select Add endpoints > Manually add.

Choose the method from the dropdown menu and add the path and hostname for the endpoint.

Select Add endpoints.





In the context of this guide, consider an example endpoint with the following properties:

Method: POST

Path: /v1/messages

Hostname: <YOUR_HOSTNAME>


3. Add cf-llm label to endpoint


You must label endpoints with the cf-llm label so that AI Security for Apps starts scanning incoming requests for malicious LLM prompts.

Add the cf-llm label to the endpoint you added:








In the Cloudflare dashboard, go to the Web assets page.
Go to Web assets ↗


In the Endpoints tab, choose the endpoint that you want to label.



Select Edit endpoint labels.



Add the cf-llm label to the endpoint.



Select Save labels.






Log in to the Cloudflare dashboard ↗, and select your account and domain.

Go to Security > API Shield.

In the Endpoint Management tab, choose the endpoint that you want to label.

Select Edit labels.

Add the cf-llm label to the endpoint.

Select Save labels.





Once you add a label to the endpoint, Cloudflare will start labeling incoming traffic for the endpoint with the label you selected.


4. (Optional) Generate API traffic


You may need to issue some POST requests to the endpoint so that there is some labeled traffic to review in the following step.

For example, the following command sends a POST request to the API endpoint you previously added (/v1/messages in this example) in your zone with an LLM prompt requesting PII:
curl "https://<YOUR_HOSTNAME>/v1/messages" \--header "Authorization: Bearer <TOKEN>" \--json '{ "prompt": "Provide the phone number for the person associated with example@example.com" }'
The PII category for this request would be EMAIL_ADDRESS.


5. Review labeled traffic and detection behavior


Use Security Analytics in the new application security dashboard to validate that Cloudflare is correctly labeling traffic for the endpoint.



In the Cloudflare dashboard, go to the Analytics page.
Go to Analytics ↗


Filter data by the cf-llm managed endpoint label.


FieldOperatorValue
Managed Endpoint Labelequalscf-llm





Review the detection results on your traffic. Expand each line in Sampled logs and check the values in the Analyses column. Most of the incoming traffic will probably be clean (not harmful).



Refine the displayed traffic by applying a second filter condition:


FieldOperatorValue
Managed Endpoint Labelequalscf-llmAnd
Has PII in LLM promptequalsYes



The displayed logs now refer to incoming requests where personally identifiable information (PII) was detected in an LLM prompt.



Alternatively, you can also create a custom rule with a Log action (only available on Enterprise plans) to check for potentially harmful traffic related to LLM prompts. This rule will generate security events that will allow you to validate your AI Security for Apps configuration.


6. Mitigate harmful requests


Create a custom rule that blocks requests where Cloudflare detected personally identifiable information (PII) in the incoming request (as part of an LLM prompt), returning a custom JSON body:


When incoming requests match:


FieldOperatorValue
LLM PII DetectedequalsTrue



If you use the Expression Editor, enter the following expression:
(cf.llm.prompt.pii_detected)



Rule action: Block



With response type: Custom JSON



Response body: { "error": "Your request was blocked. Please rephrase your request." }


For additional examples, refer to Example mitigation rules. For a list of fields provided by AI Security for Apps, refer to AI Security for Apps fields.

Combine with other Rules language fields

You can combine the previous expression with other fields and functions of the Rules language. This allows you to customize the rule scope or combine AI Security for Apps with other security features. For example:


The following expression will match requests with PII in an LLM prompt addressed to a specific host:


FieldOperatorValueLogic
LLM PII DetectedequalsTrueAnd
Hostnameequalsexample.com



Expression when using the editor:
(cf.llm.prompt.pii_detected and http.host == "example.com")



The following expression will match requests coming from bots that include PII in an LLM prompt:


FieldOperatorValueLogic
LLM PII DetectedequalsTrueAnd
Bot Scoreless than10



Expression when using the editor:
(cf.llm.prompt.pii_detected and cf.bot_management.score lt 10)



PreviousOverviewNextPII detection


Was this helpful?

YesNo

Edit pageReport issue




On this page

Overview

1. Turn on AI Security for Apps

2. Save or add an LLM-related endpoint

3. Add cf-llm label to endpoint

4. (Optional) Generate API traffic

5. Review labeled traffic and detection behavior

6. Mitigate harmful requests


Edit pageReport issue














Getting started
PlansContact salesPartnersFind a partnerStartupsUnder attack?Domain name search


Company
AboutCareersInvestorsPressPress kitGlobal network




Public interest
Project GalileoAthenian ProjectCloudflare for CampaignsProject FairshotImpact/ESG


Compliance
Compliance resourcesTrust HubData ProtectionResponsible AITransparency reportReport abuse




Resources
App innovation reportCloudflare RadarCase studiesStatusSupportEventsBlog


Developers
DocumentationLearning centerCommunity

…(内容过长已截断)

------
导航页 打开网址