Changelog

The new Snippets code editor lets you edit Snippet code and rule in one place, making it easier to test and deploy changes without switching between pages.

What’s new:

• Single-page editing for code and rule – No need to jump between screens.
• Auto-complete & syntax highlighting – Get suggestions and avoid mistakes.
• Code formatting & refactoring – Write cleaner, more readable code.

Try it now in Rules > Snippets ↗.

Hyperdrive now automatically configures your Cloudflare Tunnel to connect to your private database.

When creating a Hyperdrive configuration for a private database, you only need to provide your database credentials and set up a Cloudflare Tunnel within the private network where your database is accessible. Hyperdrive will automatically create the Cloudflare Access, Service Token, and Policies needed to secure and restrict your Cloudflare Tunnel to the Hyperdrive configuration.

To create a Hyperdrive for a private database, you can follow the Hyperdrive documentation. You can still manually create the Cloudflare Access, Service Token, and Policies if you prefer.

This feature is available from the Cloudflare dashboard.

You can now have up to 1000 Workers KV namespaces per account.

Workers KV namespace limits were increased from 200 to 1000 for all accounts. Higher limits for Workers KV namespaces enable better organization of key-value data, such as by category, tenant, or environment.

Consult the Workers KV limits documentation for the rest of the limits. This increased limit is available for both the Free and Paid Workers plans.

When using a Worker with the nodejs_compat compatibility flag enabled, you can now use the following Node.js APIs:

• node:net
• node:dns
• node:timers

You can use node:net ↗ to create a direct connection to servers via a TCP sockets with net.Socket ↗.

import net from "node:net";

const exampleIP = "127.0.0.1";

export default {
	async fetch(req) {
		const socket = new net.Socket();
		socket.connect(4000, exampleIP, function () {
			console.log("Connected");
		});

		socket.write("Hello, Server!");
		socket.end();

		return new Response("Wrote to server", { status: 200 });
	},
};

import net from "node:net";

const exampleIP = "127.0.0.1";

export default {
  async fetch(req): Promise<Response> {
    const socket = new net.Socket();
    socket.connect(4000, exampleIP, function () {
      console.log("Connected");
    });

    socket.write("Hello, Server!");
    socket.end();

    return new Response("Wrote to server", { status: 200 });
  },
} satisfies ExportedHandler;

Additionally, you can now use other APIs including net.BlockList ↗ and net.SocketAddress ↗.

Note that net.Server ↗ is not supported.

You can use node:dns ↗ for name resolution via DNS over HTTPS using Cloudflare DNS ↗ at 1.1.1.1.

import dns from "node:dns";

let response = await dns.promises.resolve4("cloudflare.com", "NS");

import dns from 'node:dns';

let response = await dns.promises.resolve4('cloudflare.com', 'NS');

All node:dns functions are available, except lookup, lookupService, and resolve which throw "Not implemented" errors when called.

You can use node:timers ↗ to schedule functions to be called at some future period of time.

This includes setTimeout ↗ for calling a function after a delay, setInterval ↗ for calling a function repeatedly, and setImmediate ↗ for calling a function in the next iteration of the event loop.

import timers from "node:timers";

console.log("first");
timers.setTimeout(() => {
	console.log("last");
}, 10);

timers.setTimeout(() => {
	console.log("next");
});

import timers from "node:timers";

console.log("first");
timers.setTimeout(() => {
  console.log("last");
}, 10);

timers.setTimeout(() => {
  console.log("next");
});

You can now detect source code leaks with Data Loss Prevention (DLP) with predefined checks against common programming languages.

The following programming languages are validated with natural language processing (NLP).

• C
• C++
• C#
• Go
• Haskell
• Java
• JavaScript
• Lua
• Python
• R
• Rust
• Swift

DLP also supports confidence level for source code profiles.

For more details, refer to DLP profiles.

Cloudflare now allows you to send SSH command logs to storage destinations configured in Logpush, including third-party destinations. Once exported, analyze and audit the data as best fits your organization! For a list of available data fields, refer to the SSH logs dataset.

To set up a Logpush job, refer to Logpush integration.

Workflows (beta) now allows you to define up to 1024 steps. sleep steps do not count against this limit.

We've also added:

• instanceId as property to the WorkflowEvent type, allowing you to retrieve the current instance ID from within a running Workflow instance
• Improved queueing logic for Workflow instances beyond the current maximum concurrent instances, reducing the cases where instances are stuck in the queued state.
• Support for pause and resume for Workflow instances in a queued state.

We're continuing to work on increases to the number of concurrent Workflow instances, steps, and support for a new waitForEvent API over the coming weeks.

Rules Overview gives you a single page to manage all your Cloudflare Rules.

What you can do:

• See all your rules in one place – No more clicking around.
• Find rules faster – Search by name.
• Understand execution order – See how rules run in sequence.
• Debug easily – Use Trace without switching tabs.

Check it out in Rules > Overview ↗.

You can now achieve higher cache hit rates and reduce origin load when using Load Balancing with Smart Tiered Cache. Cloudflare automatically selects a single, optimal tiered data center for all origins in your Load Balancing Pool.

When you use Load Balancing with Smart Tiered Cache, Cloudflare analyzes performance metrics across your pool's origins and automatically selects the optimal Upper Tier data center for the entire pool. This means:

• Consistent cache location: All origins in the pool share the same Upper Tier cache.
• Higher HIT rates: Requests for the same content hit the cache more frequently.
• Reduced origin requests: Fewer requests reach your origin servers.
• Improved performance: Faster response times for cache HITs.

Load Balancing Pool: api-pool
├── Origin 1: api-1.example.com
├── Origin 2: api-2.example.com
└── Origin 3: api-3.example.com
    ↓
Selected Upper Tier: [Optimal data center based on pool performance]

To get started, enable Smart Tiered Cache on your zone and configure your Load Balancing Pool.

Users making D1 requests via the Workers API can see up to a 60% end-to-end latency improvement due to the removal of redundant network round trips needed for each request to a D1 database.

p50, p90, and p95 request latency aggregated across entire D1 service. These latencies are a reference point and should not be viewed as your exact workload improvement.

This performance improvement benefits all D1 Worker API traffic, especially cross-region requests where network latency is an outsized latency factor. For example, a user in Europe talking to a database in North America. D1 location hints can be used to influence the geographic location of a database.

For more details on how D1 removed redundant round trips, see the D1 specific release note entry.

AI Gateway now supports DeepSeek, including their cutting-edge DeepSeek-V3 model. With this addition, you have even more flexibility to manage and optimize your AI workloads using AI Gateway. Whether you're leveraging DeepSeek or other providers, like OpenAI, Anthropic, or Workers AI, AI Gateway empowers you to:

• Monitor: Gain actionable insights with analytics and logs.
• Control: Implement caching, rate limiting, and fallbacks.
• Optimize: Improve performance with feedback and evaluations.

To get started, simply update the base URL of your DeepSeek API calls to route through AI Gateway. Here's how you can send a request using cURL:

curl https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id}/deepseek/chat/completions \
 --header 'content-type: application/json' \
 --header 'Authorization: Bearer DEEPSEEK_TOKEN' \
 --data '{
    "model": "deepseek-chat",
    "messages": [
        {
            "role": "user",
            "content": "What is Cloudflare?"
        }
    ]
}'

For detailed setup instructions, see our DeepSeek provider documentation.

Workers Builds, the integrated CI/CD system for Workers (currently in beta), now lets you cache artifacts across builds, speeding up build jobs by eliminating repeated work, such as downloading dependencies at the start of each build.

• Build Caching: Cache dependencies and build outputs between builds with a shared project-wide cache, ensuring faster builds for the entire team.

• Build Watch Paths: Define paths to include or exclude from the build process, ideal for monorepos to target only the files that need to be rebuilt per Workers project.

To get started, select your Worker on the Cloudflare dashboard ↗ then go to Settings > Builds, and connect a GitHub or GitLab repository. Once connected, you'll see options to configure Build Caching and Build Watch Paths.

After you triage your users' submissions (that are machine reviewed), you can now escalate them to our team for reclassification (which are instead human reviewed). User submissions from the submission alias, PhishNet, and our API can all be escalated.

From Reclassifications, go to User submissions. Select the three dots next to any of the user submissions, then select Escalate to create a team request for reclassification. The Cloudflare dashboard will then show you the submissions on the Team Submissions tab.

Refer to User submissions to learn more about this feature.

This feature is available across these Email security packages:

• Advantage
• Enterprise
• Enterprise + PhishGuard

You now have more transparency about team and user submissions for phishing emails through a Reclassification tab in the Zero Trust dashboard.

Reclassifications happen when users or admins submit a phish to Email security. Cloudflare reviews and - in some cases - reclassifies these emails based on improvements to our machine learning models.

This new tab increases your visibility into this process, allowing you to view what submissions you have made and what the outcomes of those submissions are.

The latest cloudflared build 2024.12.2 ↗ introduces the ability to collect all the diagnostic logs needed to troubleshoot a cloudflared instance.

A diagnostic report collects data from a single instance of cloudflared running on the local machine and outputs it to a cloudflared-diag file.

For more information, refer to Diagnostic logs.

Magic WAN and Magic Transit customers can use the Cloudflare dashboard to configure and manage BGP peering between their networks and their Magic routing table when using a Direct CNI on-ramp.

Using BGP peering allows customers to:

• Automate the process of adding or removing networks and subnets.
• Take advantage of failure detection and session recovery features.

With this functionality, customers can:

• Establish an eBGP session between their devices and the Magic WAN / Magic Transit service when connected via CNI.
• Secure the session by MD5 authentication to prevent misconfigurations.
• Exchange routes dynamically between their devices and their Magic routing table.

Refer to Magic WAN BGP peering or Magic Transit BGP peering to learn more about this feature and how to set it up.

Hyperdrive now caches queries in all Cloudflare locations, decreasing cache hit latency by up to 90%.

When you make a query to your database and Hyperdrive has cached the query results, Hyperdrive will now return the results from the nearest cache. By caching data closer to your users, the latency for cache hits reduces by up to 90%.

This reduction in cache hit latency is reflected in a reduction of the session duration for all queries (cached and uncached) from Cloudflare Workers to Hyperdrive, as illustrated below.

P50, P75, and P90 Hyperdrive session latency for all client connection sessions (both cached and uncached queries) for Hyperdrive configurations with caching enabled during the rollout period.

This performance improvement is applied to all new and existing Hyperdrive configurations that have caching enabled.

For more details on how Hyperdrive performs query caching, refer to the Hyperdrive documentation.

Now, you can manage Cloudflare Snippets with Terraform. Use infrastructure-as-code to deploy and update Snippet code and rules without manual changes in the dashboard.

Example Terraform configuration:

resource "cloudflare_snippet" "my_snippet" {
	zone_id  = "<ZONE_ID>"
	name = "my_test_snippet_1"
	main_module = "file1.js"
	files {
		name = "file1.js"
		content = file("file1.js")
	}
}

resource "cloudflare_snippet_rules" "cookie_snippet_rule" {
	zone_id  = "<ZONE_ID>"
	rules {
		enabled = true
		expression = "http.cookie eq \"a=b\""
		description = "Trigger snippet on specific cookie"
		snippet_name = "my_test_snippet_1"
	}
	depends_on = [cloudflare_snippet.my_snippet]
}

Learn more in the Configure Snippets using Terraform documentation.

You can now generate customized terraform files for building cloud network on-ramps to Magic WAN.

Magic Cloud can scan and discover existing network resources and generate the required terraform files to automate cloud resource deployment using their existing infrastructure-as-code workflows for cloud automation.

You might want to do this to:

• Review the proposed configuration for an on-ramp before deploying it with Cloudflare.
• Deploy the on-ramp using your own infrastructure-as-code pipeline instead of deploying it with Cloudflare.

For more details, refer to Set up with Terraform.

You can now use CASB to find security misconfigurations in your AWS cloud environment using Data Loss Prevention.

You can also connect your AWS compute account to extract and scan your S3 buckets for sensitive data while avoiding egress fees. CASB will scan any objects that exist in the bucket at the time of configuration.

To connect a compute account to your AWS integration:

1. In Cloudflare One ↗, go to Cloud & SaaS findings > Integrations.
2. Find and select your AWS integration.
3. Select Open connection instructions.
4. Follow the instructions provided to connect a new compute account.
5. Select Refresh.

Now, you can use Cloud Connector to route traffic to your R2 buckets based on URLs, headers, geolocation, and more.

Example setup:

curl --request PUT \
"https://api.cloudflare.com/client/v4/zones/{zone_id}/cloud_connector/rules" \
--header "Authorization: Bearer <API_TOKEN>" \
--header "Content-Type: application/json" \
--data '[
  {
    "expression": "http.request.uri.path wildcard \"/images/*\"",
    "provider": "cloudflare_r2",
    "description": "Connect to R2 bucket containing images",
    "parameters": {
      "host": "mybucketcustomdomain.example.com"
    }
  }
]'

Get started using Cloud Connector documentation.

You can now type in languages that use diacritics (like á or ç) and character-based scripts (such as Chinese, Japanese, and Korean) directly within the remote browser. The isolated browser now properly recognizes non-English keyboard input, eliminating the need to copy and paste content from a local browser or device.