Changelog

Internal DNS is now in open beta.

Internal DNS is bundled as a part of Cloudflare Gateway and is now available to every Enterprise customer with one of the following subscriptions:

• Cloudflare Zero Trust Enterprise
• Cloudflare Gateway Enterprise

To learn more and get started, refer to the Internal DNS documentation.

This week's release introduces new detections for a critical authentication bypass vulnerability in Fortinet products (CVE-2025-59718), alongside three new generic detection rules designed to identify and block HTTP Parameter Pollution attempts. Additionally, this release includes targeted protection for a high-impact unrestricted file upload vulnerability in Magento and Adobe Commerce.

Key Findings

• CVE-2025-59718: An improper cryptographic signature verification vulnerability in Fortinet FortiOS, FortiProxy, and FortiSwitchManager. This may allow an unauthenticated attacker to bypass the FortiCloud SSO login authentication using a maliciously crafted SAML message, if that feature is enabled on the device.

• Magento 2 - Unrestricted File Upload: A critical flaw in Magento and Adobe Commerce allows unauthenticated attackers to bypass security checks and upload malicious files to the server, potentially leading to Remote Code Execution (RCE).

Impact

Successful exploitation of the Fortinet and Magento vulnerabilities could allow unauthenticated attackers to gain administrative control or deploy webshells, leading to complete server compromise and data theft.

Four new fields are now available on request.cf.tlsClientAuth in Workers for requests that include a mutual TLS (mTLS) client certificate. These fields encode the client certificate and its intermediate chain in RFC 9440 ↗ format — the same standard format used by the Client-Cert and Client-Cert-Chain HTTP headers — so your Worker can forward them directly to your origin without any custom parsing or encoding logic.

export default {
  async fetch(request) {
    const tls = request.cf.tlsClientAuth;

    // Only forward if cert was verified and chain is complete
    if (!tls || !tls.certVerified || tls.certRevoked || tls.certChainRFC9440TooLarge) {
      return new Response("Unauthorized", { status: 401 });
    }

    const headers = new Headers(request.headers);
    headers.set("Client-Cert", tls.certRFC9440);
    headers.set("Client-Cert-Chain", tls.certChainRFC9440);

    return fetch(new Request(request, { headers }));
  },
};

For more information, refer to Client certificate variables and Mutual TLS authentication.

MCP server portals support Code Mode MCP server patterns, a technique that reduces context window usage by replacing individual tool definitions with a single code execution tool. Code Mode is turned on by default on all portals.

To turn it off, edit the portal in Access controls > AI controls and turn off Code Mode under Basic information.

When Code Mode is active, the portal exposes a single code tool instead of listing every tool from every upstream MCP server. The connected AI agent writes JavaScript that calls typed codemode.* methods for each upstream tool. The generated code runs in an isolated Dynamic Worker environment, keeping authentication credentials and environment variables out of the model context.

To use Code Mode, append ?codemode=search_and_execute to your portal URL when connecting from an MCP client:

https://<subdomain>.<domain>/mcp?codemode=search_and_execute

For more information, refer to Code Mode.

MCP server portals support two context optimization options that reduce how many tokens tool definitions consume in the model's context window. Both options are activated by appending the optimize_context query parameter to the portal URL.

Strips tool descriptions and input schemas from all upstream tools, leaving only their names. The portal exposes a special query tool that agents use to retrieve full definitions on demand. This provides up to 5x savings in token usage.

https://<subdomain>.<domain>/mcp?optimize_context=minimize_tools

Hides all upstream tools and exposes only two tools: query and execute. The query tool searches and retrieves tool definitions. The execute tool runs the upstream tools in an isolated Dynamic Worker environment. This reduces the initial token cost to a small constant, regardless of how many tools are available through the portal.

https://<subdomain>.<domain>/mcp?optimize_context=search_and_execute

For more information, refer to Optimize context.

Containers and Sandboxes now support connecting directly to Workers over HTTP. This allows you to call Workers functions and bindings, like KV or R2, from within the container at specific hostnames.

Define an outbound handler to capture any HTTP request or use outboundByHost to capture requests to individual hostnames and IPs.

export class MyApp extends Sandbox {}

MyApp.outbound = async (request, env, ctx) => {
	// you can run arbitrary functions defined in your Worker on any HTTP request
	return await someWorkersFunction(request.body);
};

MyApp.outboundByHost = {
	"my.worker": async (request, env, ctx) => {
		return await anotherFunction(request.body);
	},
};

In this example, requests from the container to http://my.worker will run the function defined within outboundByHost, and any other HTTP requests will run the outbound handler. These handlers run entirely inside the Workers runtime, outside of the container sandbox.

Each handler has access to env, so it can call any binding set in Wrangler config. Code inside the container makes a standard HTTP request to that hostname and the outbound Worker translates it into a binding call.

export class MyApp extends Sandbox {}

MyApp.outboundByHost = {
	"my.kv": async (request, env, ctx) => {
		const key = new URL(request.url).pathname.slice(1);
		const value = await env.KV.get(key);
		return new Response(value ?? "", { status: value ? 200 : 404 });
	},
	"my.r2": async (request, env, ctx) => {
		const key = new URL(request.url).pathname.slice(1);
		const object = await env.BUCKET.get(key);
		return new Response(object?.body ?? "", { status: object ? 200 : 404 });
	},
};

Now, from inside the container sandbox, curl http://my.kv/some-key will access Workers KV and curl http://my.r2/some-object will access R2.

Use ctx.containerId to reference the container's automatically provisioned Durable Object.

export class MyContainer extends Container {}

MyContainer.outboundByHost = {
	"get-state.do": async (request, env, ctx) => {
		const id = env.MY_CONTAINER.idFromString(ctx.containerId);
		const stub = env.MY_CONTAINER.get(id);
		return stub.getStateForKey(request.body);
	},
};

This provides an easy way to associate state with any container instance, and includes a built-in SQLite database.

Upgrade to @cloudflare/containers version 0.2.0 or later, or @cloudflare/sandbox version 0.8.0 or later to use outbound Workers.

Refer to Containers outbound traffic and Sandboxes outbound traffic for more details and examples.

DLP now processes ZIP files using a streaming handler that scans archive contents element-by-element as data arrives. This removes previous file size limitations and improves memory efficiency when scanning large archives.

Microsoft Office documents (DOCX, XLSX, PPTX) also benefit from this improvement, as they use ZIP as a container format.

This improvement is automatic — no configuration changes are required.

ctx.id.jurisdiction inside a Durable Object now reports the jurisdiction the object was created in — for example "eu" when accessed through env.MY_DURABLE_OBJECT.jurisdiction("eu") — so you can make region-aware decisions without passing the jurisdiction through method arguments or persisting it in storage. For the full list of ID-construction paths that preserve jurisdiction, refer to the Durable Object ID documentation.

export class RegionalRoom extends DurableObject {
	async fetch(request) {
		// "eu" when accessed through env.MY_DURABLE_OBJECT.jurisdiction("eu")
		const region = this.ctx.id.jurisdiction;
		return new Response(`Hello from ${region ?? "the default region"}!`);
	}
}

// Worker
export default {
	async fetch(request, env) {
		const stub = env.MY_DURABLE_OBJECT.jurisdiction("eu").getByName("general");
		return stub.fetch(request);
	},
};

ctx.id.jurisdiction is undefined for Durable Objects that were not created in a jurisdiction-restricted namespace. Alarms scheduled before 2026-03-15 also do not have jurisdiction stored; to backfill the value, reschedule the alarm from a fetch() or RPC handler.

Radar ships several improvements to the URL Scanner ↗ that make scan reports more informative and easier to share:

• Live screenshots — the summary card now includes an option to capture a live screenshot of the scanned URL on demand using the Browser Rendering API.
• Save as PDF — a new button generates a print-optimized document aggregating all tab contents (Summary, Security, Network, Behavior, and Indicators) into a single file.
• Download as JSON — raw scan data is available as a JSON download for programmatic use.
• Redesigned summary layout — page information and security details are now displayed side by side with the screenshot, with a layout that adapts to narrower viewports.
• File downloads — downloads are separated into a dedicated card with expandable rows showing each file's source URL and SHA256 hash.
• Detailed IP address data — the Network tab now includes additional detail per IP address observed during the scan.

Explore these improvements on the Cloudflare Radar URL Scanner ↗.

HTTP Archive (HAR) files are used by engineering and support teams to capture and share web traffic logs for troubleshooting. However, these files routinely contain highly sensitive data — including session cookies, authorization headers, and other credentials — that can pose a significant risk if uploaded to third-party services without being reviewed or cleaned first.

Gateway now includes a predefined DLP profile called Unsanitized HAR that detects HAR files in HTTP traffic. You can use this profile in a Gateway HTTP policy to either block HAR file uploads entirely or redirect users to a sanitization tool before allowing the upload to proceed.

In the Cloudflare dashboard ↗, go to Zero Trust > Traffic policies > Firewall Policies > HTTP and create a new HTTP policy using the DLP Profile selector:

Then choose one of the following actions:

• Block: Prevents the upload of any HAR file that has not been sanitized by Cloudflare's sanitizer. Use this for strict environments where HAR file sharing must be disallowed entirely.
• Block with Gateway Redirect: Intercepts the upload and redirects the user to https://har-sanitizer.pages.dev/, where they can sanitize the file. Once sanitized, the user can re-upload the clean file and proceed with their workflow.

HAR files processed by the Cloudflare HAR sanitizer receive a tamper-evident sanitized marker. DLP recognizes this marker and will not re-trigger the policy on a file that has already been sanitized and has not been modified since. If a previously sanitized file is edited, it will be treated as unsanitized and flagged again.

Gateway logs will reflect whether a detected HAR file was classified as Unsanitized or Sanitized, giving your security team full visibility into HAR file activity across your organization.

For more information, refer to predefined DLP profiles.

Logpush now supports higher-precision timestamp formats for log output. You can configure jobs to output timestamps at millisecond or nanosecond precision. This is available in both the Logpush UI in the Cloudflare dashboard and the Logpush API.

To use the new formats, set timestamp_format in your Logpush job's output_options:

• rfc3339ms — 2024-02-17T23:52:01.123Z
• rfc3339ns — 2024-02-17T23:52:01.123456789Z

Default timestamp formats apply unless explicitly set. The dashboard defaults to rfc3339 and the API defaults to unixnano.

For more information, refer to the Log output options documentation.

Cloudflare now exposes four new fields in the Transform Rules phase that encode client certificate data in RFC 9440 ↗ format. Previously, forwarding client certificate information to your origin required custom parsing of PEM-encoded fields or non-standard HTTP header formats. These new fields produce output in the standardized Client-Cert and Client-Cert-Chain header format defined by RFC 9440, so your origin can consume them directly without any additional decoding logic.

Each certificate is DER-encoded, Base64-encoded, and wrapped in colons. For example, :MIIDsT...Vw==:. A chain of intermediates is expressed as a comma-separated list of such values.

The chain encoding follows the same ordering as the TLS handshake: the certificate closest to the leaf appears first, working up toward the trust anchor. The root certificate is not included.

Add a request header transform rule to set the Client-Cert and Client-Cert-Chain headers on requests forwarded to your origin server. For example, to forward headers for verified, non-revoked certificates:

Rule expression:

cf.tls_client_auth.cert_verified and not cf.tls_client_auth.cert_revoked

Header modifications:

To get the most out of these fields, upload your client CA certificate to Cloudflare so that Cloudflare validates the client certificate at the edge and populates cf.tls_client_auth.cert_verified and cf.tls_client_auth.cert_revoked.

For more information, refer to Mutual TLS authentication, Request Header Transform Rules, and the fields reference.

The new secrets configuration property lets you declare the secret names your Worker requires in your Wrangler configuration file. Required secrets are validated during local development and deploy, and used as the source of truth for type generation.

{
	"secrets": {
		"required": ["API_KEY", "DB_PASSWORD"],
	},
}

[secrets]
required = [ "API_KEY", "DB_PASSWORD" ]

When secrets is defined, wrangler dev and vite dev load only the keys listed in secrets.required from .dev.vars or .env/process.env. Additional keys in those files are excluded. If any required secrets are missing, a warning is logged listing the missing names.

wrangler types generates typed bindings from secrets.required instead of inferring names from .dev.vars or .env. This lets you run type generation in CI or other environments where those files are not present. Per-environment secrets are supported — the aggregated Env type marks secrets that only appear in some environments as optional.

wrangler deploy and wrangler versions upload validate that all secrets in secrets.required are configured on the Worker before the operation succeeds. If any required secrets are missing, the command fails with an error listing which secrets need to be set.

For more information, refer to the secrets configuration property reference.

AI Crawl Control now supports extending the underlying WAF rule with custom modifications. Any changes you make directly in the WAF custom rules editor — such as adding path-based exceptions, extra user agents, or additional expression clauses — are preserved when you update crawler actions in AI Crawl Control.

If the WAF rule expression has been modified in a way AI Crawl Control cannot parse, a warning banner appears on the Crawlers page with a link to view the rule directly in WAF.

For more information, refer to WAF rule management.

You can now control how Cloudflare handles origin responses without changing your origin. Cache Response Rules let you modify Cache-Control directives, manage cache tags, and strip headers like Set-Cookie from origin responses before they reach Cloudflare's cache. Whether traffic is cached or passed through dynamically, these rules give you control over origin response behavior that was previously out of reach.

Cache Rules previously only operated on request attributes. Cache Response Rules introduce a new response phase that evaluates origin responses and lets you act on them before caching. You can now:

• Modify Cache-Control directives: Set or remove individual directives like no-store, no-cache, max-age, s-maxage, stale-while-revalidate, immutable, and more. For example, remove a no-cache directive your origin sends so Cloudflare can cache the asset, or set an s-maxage to control how long Cloudflare stores it.
• Set a different browser Cache-Control: Send a different Cache-Control header downstream to browsers and other clients than what Cloudflare uses internally, giving you independent control over edge and browser caching strategies.
• Manage cache tags: Add, set, or remove cache tags on responses, including converting tags from another CDN's header format into Cloudflare's Cache-Tag header. This is especially useful if you are migrating from a CDN that uses a different tag header or delimiter.
• Strip headers that block caching: Remove Set-Cookie, ETag, or Last-Modified headers from origin responses before caching, so responses that would otherwise be treated as uncacheable can be stored and served from cache.

• No origin changes required: Fix caching behavior entirely from Cloudflare, even when your origin configuration is locked down or managed by a different team.
• Simpler CDN migration: Match caching behavior from other CDN providers without rewriting your origin. Translate cache tag formats and override directives that do not align with Cloudflare's defaults.
• Native support, fewer workarounds: Functionality that previously required workarounds is now built into Cache Rules with full Tiered Cache compatibility.
• Fine-grained control: Use expressions to match on request and response attributes, then apply precise cache settings per rule. Rules are stackable and composable with existing Cache Rules.

Configure Cache Response Rules in the Cloudflare dashboard ↗ under Caching > Cache Rules, or via the Rulesets API ↗. For more details, refer to the Cache Rules documentation ↗.

Containers now support Docker Hub ↗ images. You can use a fully qualified Docker Hub image reference in your Wrangler configuration ↗ instead of first pushing the image to Cloudflare Registry.

{
	"containers": [
		{
			// Example: docker.io/cloudflare/sandbox:0.7.18
			"image": "docker.io/<NAMESPACE>/<REPOSITORY>:<TAG>",
		},
	],
}

[[containers]]
image = "docker.io/<NAMESPACE>/<REPOSITORY>:<TAG>"

Containers also support private Docker Hub images. To configure credentials, refer to Use private Docker Hub images.

For more information, refer to Image management.

Cloudflare Gateway now supports OIDC Claims as a selector in Firewall, Resolver, and Egress policies. Administrators can use custom OIDC claims from their identity provider to build fine-grained, identity-based traffic policies across all Gateway policy types.

With this update, you can:

• Filter traffic in DNS, HTTP, and Network firewall policies based on OIDC claim values.
• Apply custom resolver policies to route DNS queries to specific resolvers depending on a user's OIDC claims.
• Control egress policies to assign dedicated egress IPs based on OIDC claim attributes.

For example, you can create a policy that routes traffic differently for users with department=engineering in their OIDC claims, or restrict access to certain destinations based on a user's role claim.

To get started, configure custom OIDC claims on your identity provider and use the OIDC Claims selector in the Gateway policy builder.

For more information, refer to Identity-based policies.

The top-level Interconnects page in the Cloudflare dashboard has been removed. Interconnects are now located under Connectors > Interconnects.

Your existing configurations and functionality remain the same.

Dynamic Workers are now in open beta ↗ for all paid Workers users. You can now have a Worker spin up other Workers, called Dynamic Workers, at runtime to execute code on-demand in a secure, sandboxed environment. Dynamic Workers start in milliseconds, making them well suited for fast, secure code execution at scale.

• Code Mode: LLMs are trained to write code. Run tool-calling logic written in code instead of stepping through many tool calls, which can save up to 80% in inference tokens and cost.
• AI agents executing code: Run code for tasks like data analysis, file transformation, API calls, and chained actions.
• Running AI-generated code: Run generated code for prototypes, projects, and automations in a secure, isolated sandboxed environment.
• Fast development and previews: Load prototypes, previews, and playgrounds in milliseconds.
• Custom automations: Create custom tools on the fly that execute a task, call an integration, or automate a workflow.

Dynamic Workers support two loading modes:

• load(code) — for one-time code execution (equivalent to calling get() with a null ID).
• get(id, callback) — caches a Dynamic Worker by ID so it can stay warm across requests. Use this when the same code will receive subsequent requests.

export default {
	async fetch(request, env) {
		const worker = env.LOADER.load({
			compatibilityDate: "2026-01-01",
			mainModule: "src/index.js",
			modules: {
				"src/index.js": `
					export default {
						fetch() {
							return new Response("Hello from a dynamic Worker");
						},
					};
				`,
			},
			// Block all outbound network access from the Dynamic Worker.
			globalOutbound: null,
		});

		return worker.getEntrypoint().fetch(request);
	},
};

export default {
	async fetch(request: Request, env: Env): Promise<Response> {
		const worker = env.LOADER.load({
			compatibilityDate: "2026-01-01",
			mainModule: "src/index.js",
			modules: {
				"src/index.js": `
					export default {
						fetch() {
							return new Response("Hello from a dynamic Worker");
						},
					};
				`,
			},
			// Block all outbound network access from the Dynamic Worker.
			globalOutbound: null,
		});

		return worker.getEntrypoint().fetch(request);
	},
};

Here are 3 new libraries to help you build with Dynamic Workers:

• @cloudflare/codemode ↗: Replace individual tool calls with a single code() tool, so LLMs write and execute TypeScript that orchestrates multiple API calls in one pass.

• @cloudflare/worker-bundler ↗: Resolve npm dependencies and bundle source files into ready-to-load modules for Dynamic Workers, all at runtime.

• @cloudflare/shell ↗: Give your agent a virtual filesystem inside a Dynamic Worker with persistent storage backed by SQLite and R2.

Dynamic Workers Starter

Use this starter ↗ to deploy a Worker that can load and execute Dynamic Workers.

Dynamic Workers Playground

Deploy the Dynamic Workers Playground ↗ to write or import code, bundle it at runtime with @cloudflare/worker-bundler, execute it through a Dynamic Worker, and see real-time responses and execution logs.

For the full API reference and configuration options, refer to the Dynamic Workers documentation.

Dynamic Workers pricing is based on three dimensions: Dynamic Workers created daily, requests, and CPU time.

¹ Uses Workers Standard rates and will appear as part of your existing Workers bill, not as separate Dynamic Workers charges.

Note: Dynamic Workers requests and CPU time are already billed as part of your Workers plan and will count toward your Workers requests and CPU usage. The Dynamic Workers created daily charge is not yet active — you will not be billed for the number of Dynamic Workers created at this time. Pricing information is shared in advance so you can estimate future costs.

Workflow instance methods pause(), resume(), restart(), and terminate() are now available in local development when using wrangler dev.

You can now test the full Workflow instance lifecycle locally:

const instance = await env.MY_WORKFLOW.create({
	id: "my-instance-id",
});

await instance.pause(); // pauses a running workflow instance
await instance.resume(); // resumes a paused instance
await instance.restart(); // restarts the instance from the beginning
await instance.terminate(); // terminates the instance immediately

The latest release of the Agents SDK ↗ exposes agent state as a readable property, prevents duplicate schedule rows across Durable Object restarts, brings full TypeScript inference to AgentClient, and migrates to Zod 4.

Both useAgent (React) and AgentClient (vanilla JS) now expose a state property that reflects the current agent state. Previously, reading state required manually tracking it through the onStateUpdate callback.

React (useAgent)

const agent = useAgent({
	agent: "game-agent",
	name: "room-123",
});

// Read state directly — no separate useState + onStateUpdate needed
return <div>Score: {agent.state?.score}</div>;

// Spread for partial updates
agent.setState({ ...agent.state, score: (agent.state?.score ?? 0) + 10 });

const agent = useAgent<GameAgent, GameState>({
	agent: "game-agent",
	name: "room-123",
});

// Read state directly — no separate useState + onStateUpdate needed
return <div>Score: {agent.state?.score}</div>;

// Spread for partial updates
agent.setState({ ...agent.state, score: (agent.state?.score ?? 0) + 10 });

agent.state is reactive — the component re-renders when state changes from either the server or a client-side setState() call.

Vanilla JS (AgentClient)

const client = new AgentClient({
	agent: "game-agent",
	name: "room-123",
	host: "your-worker.workers.dev",
});

client.setState({ score: 100 });
console.log(client.state); // { score: 100 }

const client = new AgentClient<GameAgent>({
	agent: "game-agent",
	name: "room-123",
	host: "your-worker.workers.dev",
});

client.setState({ score: 100 });
console.log(client.state); // { score: 100 }

State starts as undefined and is populated when the server sends the initial state on connect (from initialState) or when setState() is called. Use optional chaining (agent.state?.field) for safe access. The onStateUpdate callback continues to work as before — the new state property is additive.

schedule() now supports an idempotent option that deduplicates by (type, callback, payload), preventing duplicate rows from accumulating when called in places that run on every Durable Object restart such as onStart().

Cron schedules are idempotent by default. Calling schedule("0 * * * *", "tick") multiple times with the same callback, expression, and payload returns the existing schedule row instead of creating a new one. Pass { idempotent: false } to override.

Delayed and date-scheduled types support opt-in idempotency:

import { Agent } from "agents";

class MyAgent extends Agent {
	async onStart() {
		// Safe across restarts — only one row is created
		await this.schedule(60, "maintenance", undefined, { idempotent: true });
	}
}

import { Agent } from "agents";

class MyAgent extends Agent {
	async onStart() {
		// Safe across restarts — only one row is created
		await this.schedule(60, "maintenance", undefined, { idempotent: true });
	}
}

Two new warnings help catch common foot-guns:

• Calling schedule() inside onStart() without { idempotent: true } emits a console.warn with actionable guidance (once per callback; skipped for cron and when idempotent is set explicitly).
• If an alarm cycle processes 10 or more stale one-shot rows for the same callback, the SDK emits a console.warn and a schedule:duplicate_warning diagnostics channel event.

AgentClient now accepts an optional agent type parameter for full type inference on RPC calls, matching the typed experience already available with useAgent.

const client = new AgentClient({
	agent: "my-agent",
	host: window.location.host,
});

// Typed call — method name autocompletes, args and return type inferred
const value = await client.call("getValue");

// Typed stub — direct RPC-style proxy
await client.stub.getValue();
await client.stub.add(1, 2);

const client = new AgentClient<MyAgent>({
	agent: "my-agent",
	host: window.location.host,
});

// Typed call — method name autocompletes, args and return type inferred
const value = await client.call("getValue");

// Typed stub — direct RPC-style proxy
await client.stub.getValue();
await client.stub.add(1, 2);

State is automatically inferred from the agent type, so onStateUpdate is also typed:

const client = new AgentClient({
	agent: "my-agent",
	host: window.location.host,
	onStateUpdate: (state) => {
		// state is typed as MyAgent's state type
	},
});

const client = new AgentClient<MyAgent>({
	agent: "my-agent",
	host: window.location.host,
	onStateUpdate: (state) => {
		// state is typed as MyAgent's state type
	},
});

Existing untyped usage continues to work without changes. The RPC type utilities (AgentMethods, AgentStub, RPCMethods) are now exported from agents/client for advanced typing scenarios. agents, @cloudflare/ai-chat, and @cloudflare/codemode now require zod ^4.0.0. Zod v3 is no longer supported.

• Turn serialization — onChatMessage() and _reply() work is now queued so user requests, tool continuations, and saveMessages() never stream concurrently.
• Duplicate messages on stop — Clicking stop during an active stream no longer splits the assistant message into two entries.
• Duplicate messages after tool calls — Orphaned client IDs no longer leak into persistent storage.

keepAlive() now uses a lightweight in-memory ref count instead of schedule rows. Multiple concurrent callers share a single alarm cycle. The @experimental tag has been removed from both keepAlive() and keepAliveWhile().

A new entry point @cloudflare/codemode/tanstack-ai adds support for TanStack AI's ↗ chat() as an alternative to the Vercel AI SDK's streamText():

import {
	createCodeTool,
	tanstackTools,
} from "@cloudflare/codemode/tanstack-ai";
import { chat } from "@tanstack/ai";

const codeTool = createCodeTool({
	tools: [tanstackTools(myServerTools)],
	executor,
});

const stream = chat({ adapter, tools: [codeTool], messages });

import { createCodeTool, tanstackTools } from "@cloudflare/codemode/tanstack-ai";
import { chat } from "@tanstack/ai";

const codeTool = createCodeTool({
	tools: [tanstackTools(myServerTools)],
	executor,
});

const stream = chat({ adapter, tools: [codeTool], messages });

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest

AI Search now offers new REST API endpoints for search and chat that use an OpenAI compatible format. This means you can use the familiar messages array structure that works with existing OpenAI SDKs and tools. The messages array also lets you pass previous messages within a session, so the model can maintain context across multiple turns.

Here is an example request to the Chat Completions endpoint using the new messages array format:

curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai-search/instances/{NAME}/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer {API_TOKEN}" \
  -d '{
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful documentation assistant."
      },
      {
        "role": "user",
        "content": "How do I get started?"
      }
    ]
  }'

For more details, refer to the AI Search REST API guide.

If you are using the previous AutoRAG API endpoints (/autorag/rags/), we recommend migrating to the new endpoints. The previous AutoRAG API endpoints will continue to be fully supported.

Refer to the migration guide for step-by-step instructions.

AI Search now supports public endpoints, UI snippets, and MCP, making it easy to add search to your website or connect AI agents.

Public endpoints allow you to expose AI Search capabilities without requiring API authentication. To enable public endpoints:

1. Go to AI Search in the Cloudflare dashboard. Go to AI Search ↗
2. Select your instance, and turn on Public Endpoint in Settings. For more details, refer to Public endpoint configuration.

UI snippets are pre-built search and chat components you can embed in your website. Visit search.ai.cloudflare.com ↗ to configure and preview components for your AI Search instance.

To add a search modal to your page:

<script
	type="module"
	src="https://<PUBLIC_ENDPOINT_ID>.search.ai.cloudflare.com/assets/v0.0.25/search-snippet.es.js"
></script>

<search-modal-snippet
	api-url="https://<PUBLIC_ENDPOINT_ID>.search.ai.cloudflare.com/"
	placeholder="Search..."
>
</search-modal-snippet>

For more details, refer to the UI snippets documentation.

The MCP endpoint allows AI agents to search your content via the Model Context Protocol. Connect your MCP client to:

https://<PUBLIC_ENDPOINT_ID>.search.ai.cloudflare.com/mcp

For more details, refer to the MCP documentation.

AI Search now supports custom metadata filtering, allowing you to define your own metadata fields and filter search results based on attributes like category, version, or any custom field you define.

You can define up to 5 custom metadata fields per AI Search instance. Each field has a name and data type (text, number, or boolean):

curl -X POST https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai-search/instances \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer {API_TOKEN}" \
  -d '{
    "id": "my-instance",
    "type": "r2",
    "source": "my-bucket",
    "custom_metadata": [
      { "field_name": "category", "data_type": "text" },
      { "field_name": "version", "data_type": "number" },
      { "field_name": "is_public", "data_type": "boolean" }
    ]
  }'

How you attach metadata depends on your data source:

• R2 bucket: Set metadata using S3-compatible custom headers (x-amz-meta-*) when uploading objects. Refer to R2 custom metadata for examples.
• Website: Add <meta> tags to your HTML pages. Refer to Website custom metadata for details.

Use custom metadata fields in your search queries alongside built-in attributes like folder and timestamp:

curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai-search/instances/{NAME}/search \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer {API_TOKEN}" \
  -d '{
    "messages": [
      {
        "content": "How do I configure authentication?",
        "role": "user"
      }
    ],
    "ai_search_options": {
      "retrieval": {
        "filters": {
          "category": "documentation",
          "version": { "$gte": 2.0 }
        }
      }
    }
  }'

Learn more in the metadata filtering documentation.

Two new fields are now available in the httpRequestsAdaptive and httpRequestsAdaptiveGroups GraphQL Analytics API datasets:

• webAssetsOperationId — the ID of the saved endpoint that matched the incoming request.
• webAssetsLabelsManaged — the managed labels mapped to the matched operation at the time of the request (for example, cf-llm, cf-log-in). At most 10 labels are returned per request.

Both fields are empty when no operation matched. webAssetsLabelsManaged is also empty when no managed labels are assigned to the matched operation.

These fields allow you to determine, per request, which Web Assets operation was matched and which managed labels were active. This is useful for troubleshooting downstream security detection verdicts — for example, understanding why AI Security for Apps did or did not flag a request.

Refer to Endpoint labeling service for GraphQL query examples.