OAuth client developers can now classify configured scopes as required or optional in the Cloudflare dashboard. By default, all configured scopes remain required .
What's New
Optional Scopes: OAuth clients can now mark configured scopes as optional, allowing applications to request them without requiring users to approve them.
Scope Selection: On the consent screen, users must grant required scopes but can decline optional scopes. This helps customers apply least-privilege access to applications, CLIs, and workloads. Optional scopes are selected by default.
Templates: The consent screen now includes Read Only and Full Access templates to make scope selection faster and easier.
Search: Users can now search scopes in the consent screen.
Leaked credentials detection now scans the Authorization request header for Basic Authentication credentials. Previously, the detection only inspected request bodies, query strings, and headers for well-known web applications or custom detection locations, which meant credentials sent through HTTP Basic Authentication were not covered by default.
This new default scan location decodes the Authorization: Basic <credentials> header and compares the extracted username and password against Cloudflare's database of leaked credentials, the same way as other default scan locations. Matches populate the existing leaked credentials fields, such as cf.waf.credential_check.password_leaked, and trigger the Exposed-Credential-Check managed transform header if configured, so you can reuse existing custom rules and rate limiting rules without changes.
This change was applied automatically for zones with leaked credentials detection enabled. No configuration changes are required.
This release introduces multiple features from our previous beta release into stable release, including:
When reauthentication is needed for any reason, the notifications are clearer and reduce the actions needed to get you back to work by redirecting to the browser for authentication instead of the app window when necessary.
When a network is blocking or otherwise not supportive of HTTP/3, the client will learn and adapt by switching the order of fallback for that network by starting with HTTP/2 first and then trying HTTP/3 if needed. This reduces delays in time to connectivity when joining older or heavily filtered networks.
Additional changes and improvements
Fixed a process leak in the Windows GUI that could exhaust system resources during IPC client-creation failures.
Fixed being unable to switch organizations when the client was stuck in the "Device not in organization" state.
Fixed an issue where Microsoft Defender would falsely flag the Cloudflare One Client installation as malicious when installing with Intune.
Made the Windows domain-joined posture check more reliable.
A DNS search domain parsing failure no longer prevents connection.
Cloud icon now correctly reflects actual connection status instead of showing disconnected while fully connected.
Fixed missing certificate error display due to a race condition.
Fixed empty black window after transitioning from docked dual displays to undocked/internal display.
This release introduces multiple features from our previous beta release into stable release, including:
When reauthentication is needed for any reason, the notifications are clearer and reduce the actions needed to get you back to work by redirecting to the browser for authentication instead of the app window when necessary.
When a network is blocking or otherwise not supportive of HTTP/3, the client will learn and adapt by switching the order of fallback for that network by starting with HTTP/2 first and then trying HTTP/3 if needed. This reduces delays in time to connectivity when joining older or heavily filtered networks.
Additional changes and improvements
Fixed the client not allowing login to another organization when currently showing "Device not in organization."
A DNS search domain parsing failure no longer prevents connection.
Cloud icon now correctly reflects actual connection status instead of showing disconnected while fully connected.
Fixed missing certificate error display due to a race condition.
Fixed crash when trying to connect to captive portal on Wi-Fi.
Fixed empty black window after transitioning from docked dual displays to undocked/internal display.
This release introduces multiple features from our previous beta release into stable release, including:
When reauthentication is needed for any reason, the notifications are clearer and reduce the actions needed to get you back to work by redirecting to the browser for authentication instead of the app window when necessary.
When a network is blocking or otherwise not supportive of HTTP/3, the client will learn and adapt by switching the order of fallback for that network by starting with HTTP/2 first and then trying HTTP/3 if needed. This reduces delays in time to connectivity when joining older or heavily filtered networks.
Additional changes and improvements
Fixed the client not allowing login to another organization when currently showing "Device not in organization."
A DNS search domain parsing failure no longer prevents connection.
Cloud icon now correctly reflects actual connection status instead of showing disconnected while fully connected.
Fixed missing certificate error display due to a race condition.
Fixed empty black window after transitioning from docked dual displays to undocked/internal display.
Fixed hostname routes not working for Cloudflare Mesh when the IP addresses of the hostnames are local addresses.
Known issues
When in DNS Only mode, the client may send DNS queries for names that are configured for Local Domain Fallback to the encrypted DNS server instead of falling back to the system configuration. Local Domain Fallback works as expected in other client modes.
GPT-5.6 Sol is available through AI Gateway, and for a limited time you can use it at 50% off. If you are already using AI Gateway, point to the openai/gpt-5.6-sol model and the discounted pricing applies automatically — no promo code needed.
The promotion is available for Unified Billing users only (not Bring Your Own Keys). Load credits onto AI Gateway and start sending requests to openai/gpt-5.6-sol.
Discounted pricing during the promotion:
Usage
Promotional price
Standard price
Input
$2.50 per 1M tokens
$5 per 1M tokens
Output
$15 per 1M tokens
$30 per 1M tokens
Cache read
$0.25 per 1M tokens
$0.50 per 1M tokens
The promotion runs through September 18, 2026. After that date, GPT-5.6 Sol requests return to standard pricing.
Version 1 of the Workers Vitest integration is published as @cloudflare/vitest-plugin ↗. The package was formerly named @cloudflare/vitest-pool-workers.
The Vitest configuration API is unchanged. Existing projects must update the dependency name, package imports, and TypeScript types entries.
You can now configure origin application settings directly in the Cloudflare dashboard when adding or editing a published application route for a Cloudflare Tunnel. These settings control how cloudflared connects to your origin server and were previously only available in the Cloudflare One dashboard or via local configuration files.
When editing a published application, expand Additional application settings to configure parameters organized into three categories:
HTTP — Set a custom HTTP Host header or disable chunked encoding.
TLS — Configure origin server name, CA pool, TLS timeout, disable TLS verification, match SNI to host, or enable HTTP/2 to origin.
Cloudflare Email Security now supports post-quantum hybrid key exchange with X25519MLKEM768 on the SMTP connections we make to receive and deliver mail. Deploying Email Security in front of a provider that supports post-quantum hybrid key agreement (like Google Workspace) will create a TLS 1.3 connection using post-quantum key agreement.
Inbound MX connections and outbound delivery connections now negotiate the X25519MLKEM768 hybrid key agreement when the peer supports it, protecting SMTP traffic against harvest-now, decrypt-later ↗ attacks.
Support is backwards compatible and enabled automatically for all customers. Senders and receivers that do not yet advertise post-quantum key agreement continue to connect with classical key exchange.
Load balancing analytics now filters traffic data by pool name instead of pool ID, aligning the query behavior with the pool names displayed in the filter dropdown.
Previously, the analytics pool filter queried by internal pool ID while displaying pool names in the UI dropdown. This mismatch caused filtering issues when pools shared similar names or when you expected results based on the visible pool name. Because the underlying query used a different identifier than what appeared on screen, the displayed data could be confusing or incorrect.
The pool filter now queries by the same pool name shown in the dropdown. When you select a pool from the filter, the analytics graphs and tables display data for that specific pool as you would expect. This change affects:
Requests over time, filtering the chart series to the selected pool.
Pool distribution, showing only the selected pool segment.
Top endpoints, displaying cards for origins in the selected pool.
Latency, showing latency data for the selected pool.
The Logs view and health event filtering are unchanged.
To use this, go to Traffic > Load Balancing Analytics for a zone. The same pool filter appears in the analytics view for an individual load balancer under Load Balancing at the account level.
R2 now supports a usjurisdiction, which guarantees that bucket data is stored and processed within the United States. Use this jurisdiction when you need explicit US data residency guarantees.
Use the jurisdiction-specific S3 endpoint to create and access buckets in the us jurisdiction:
https://<ACCOUNT_ID>.us.r2.cloudflarestorage.com
To access a bucket in the us jurisdiction from Workers, set jurisdiction in your R2 binding:
This release updates WordPress remote code execution rule metadata in the Cloudflare Managed Ruleset and Cloudflare Free Ruleset to identify CVE-2026-65640.
Key Findings
CVE-2026-65640: A remote code execution vulnerability affecting WordPress core and plugin components. Remote, unauthenticated attackers can execute arbitrary system commands to gain unauthorized access or establish backdoors on host servers.
Impact
The WordPress changes update rule metadata only; detection behavior and actions remain unchanged.
Qwen 3.8 27B is a 27-billion-parameter instruction-tuned vision language model from Alibaba's Qwen family. It processes images and text together, with reasoning and function calling for agentic workflows.
Key capabilities:
Vision: Accept image and text inputs and generate text responses.
Reasoning: Support thinking mode for complex, step-by-step problem-solving.
Function calling: Build agents that invoke tools and APIs across multiple conversation turns.
262,144 token context window: Retain long conversations and multimodal inputs across extended agent sessions.
Use Qwen 3.8 27B through the Workers AI binding (env.AI.run()) or the REST API at /ai/run. You can also use AI Gateway with these endpoints.
Cloudflare has fixed an issue affecting WebSocket data transfer reporting. HTTP Traffic Analytics and HTTP request logs now correctly count data transferred throughout a WebSocket connection, restoring the correct behavior. During the affected period, reporting captured only the initial 101 Switching Protocols handshake for some WebSocket connections, which could underreport their data transfer.
Customers with WebSocket traffic will see the correct Data Transfer in the dashboard and EdgeResponseBytes in analytics and HTTP request logs. The change reflects restored accounting of existing WebSocket traffic, not an increase in traffic caused by this change. WebSocket connection behavior is unaffected.
The separate WebSocket Analytics Logpush dataset continues to provide per-connection directional byte counts, timestamps, and close details.
For more information about HTTP Traffic Analytics, refer to Zone Analytics.
DeepSeek V4 Flash and DeepSeek V4 Pro are the first Workers AI models with a full one million (1,048,576) token context window. Use them for long-horizon agentic workflows, large codebases, and multi-step reasoning that exceed the context limits of every other model hosted on the platform.
DeepSeek V4 Flash is the faster, lower-cost sibling. This release supersedes the preview version with substantially enhanced agentic capabilities.
Key capabilities:
Reasoning: Both models support thinking mode for complex, step-by-step problem-solving.
Function calling: Build agents that invoke tools and APIs across multiple conversation turns.
Long context: Both models support a full 1,048,576 token context window.
Protect an application across all its domains at once
Until now, if a Worker was reachable on a route, a Custom Domain, and a workers.dev URL, you had to manually add each one to an Access application and keep the list in sync whenever routes or domains changed.
Now, Access attaches the policy to the Worker itself, so every associated domain and preview URL stays protected even when its routes or domains change.
Protect all new and existing Workers by default
Make all Workers private by default, so every existing and newly created Worker requires sign-in before anyone can reach it.
If a specific Worker should remain publicly accessible, add a Worker-level bypass to exempt it.
Whether you protect a single application or all Workers at once, you can choose whether to protect preview deployments only or both previews and production, and control who can sign in by Cloudflare account membership, email address, or email domain.
For more advanced policy options, edit the policy in Zero Trust ↗.
View all of your Worker Access policies
You can view and manage all of your Access policies in the Access tab of the Workers & Pages section in the dashboard.
See who is accessing your Worker
When Access is enabled on your Worker, every authenticated request includes ctx.access. Call ctx.access.getIdentity() to get the user's email, name, and groups — no manual JWT validation required.
export default { async fetch(request, env, ctx) { if (!ctx.access) { return new Response("Access did not run", { status: 401 }); } const identity = await ctx.access.getIdentity(); return Response.json({ aud: ctx.access.aud, email: identity?.email }); },};
Test Access locally
You can now test Cloudflare Access locally with wrangler dev. Add a dev block to your wrangler.jsonc:
Your Worker will receive this identity through ctx.access and ctx.access.getIdentity(), letting you test authenticated and unauthenticated flows without deploying. Remove the dev block to simulate unauthenticated requests.
API and programmatic access
You can also set up these policies through the Workers API instead of the dashboard.
Artifacts now supports jurisdictions, allowing you to select the European Union or the United States as the only location where repo data is stored and processed.
Select a jurisdiction when you create a namespace. Every repo in that namespace automatically uses the selected jurisdiction.
Cloudflare Gateway can now detect software package downloads and give you policy control over supply chain traffic. When a developer or CI/CD pipeline downloads a package through Gateway, the proxy identifies the registry protocol from the request URL and extracts the package ecosystem, name, version, and namespace. You can then write HTTP policies using pkg.* selectors to allow or block package downloads.
Supported ecosystems
Gateway detects package downloads for the following ecosystems:
Ecosystem
Namespace
npm
Scope (for example, @babel)
PyPI
--
RubyGems
--
Cargo
--
Go
Module path
Maven
Group ID
NuGet
--
Selectors
In the dashboard, select Package Ecosystem to access the package registry selectors. After selecting a single ecosystem, nested fields for package name, version, and namespace become available. Five pkg.* selectors are available for HTTP policies with the Allow and Block actions:
Selector
Description
pkg.ecosystem
The package ecosystem detected from the request URL.
pkg.name
The package name extracted from the download URL.
pkg.version
The package version, with support for ecosystem-aware comparison operators.
pkg.namespace
The package namespace, when the ecosystem supports one.
pkg.purl
The Package URL (PURL) ↗ derived from the detected coordinates. Available in the API only.
Detection is based on the registry protocol rather than the hostname, so it works the same way whether traffic goes to a public registry, a corporate proxy such as Artifactory or Nexus, or a self-hosted mirror.
Package registry security requires TLS decryption to be turned on.
Cloudflare Realtime SFU is a WebRTC selective forwarding unit that runs on Cloudflare's global network. It forwards audio, video, and application data between WebRTC clients without requiring you to manage SFU infrastructure or regions.
DataChannels are WebRTC channels for application messages. A client publishes a named DataChannel to Realtime SFU, and the SFU forwards its messages to every client that subscribes to that channel. Use DataChannels for low-latency payloads such as chat messages, game state, sensor updates, and control events.
What changed
Realtime SFU DataChannels now support unordered and partially reliable delivery. DataChannels remain reliable and ordered by default, so existing channels keep their current behavior.
With ordered delivery, a delayed message can block later messages. For game state or sensor updates, recent data may be more useful than recovering an older message. Unordered delivery lets later messages proceed, while partial reliability limits retransmission attempts or delivery time.
Choose delivery behavior
Delivery settings answer two questions: whether newer messages can bypass a delayed message, and when the transport should stop retrying delivery.
Choose the policy that matches how long your payload remains useful:
Goal
Settings
Use when
Reliable, ordered delivery (default)
Omit ordered, maxRetransmits, and maxPacketLifeTime
Messages remain useful and must arrive in order
Reliable, unordered delivery
Set ordered: false; omit both retry fields
Messages remain useful, but later messages should not wait for earlier messages
No retries or ordering
Set ordered: false and maxRetransmits: 0
The application tolerates message loss and discards out-of-date updates
Limited retries
Set maxRetransmits: <COUNT>
Brief recovery is useful, but repeated retries are not
Time-bounded delivery
Set maxPacketLifeTime: <MILLISECONDS>
A message loses value after a known time window
ordered controls ordering independently from retries. maxRetransmits and maxPacketLifeTime are alternative retry budgets, so set at most one for each channel. Omit both for reliable delivery, whether ordered or unordered.
Apply the policy end to end
Realtime DataChannels use negotiated IDs, so browsers do not receive delivery settings from the remote peer. Apply the same settings when the publisher creates the local channel, each subscriber pulls the remote channel, and each client calls createDataChannel().
In the publisher and subscriber clients, create the negotiated browser DataChannel with the same settings. In this example, pc is the active RTCPeerConnection, and channelId is the ID returned by the corresponding API request:
Cloud Connector now supports public Oracle Cloud Infrastructure (OCI) Object Storage buckets. You can route matching requests to OCI without managing a separate origin-routing configuration.
OCI support uses the Amazon S3 Compatibility API. Both path-style and virtual-hosted endpoint formats are supported, including traditional oraclecloud.com and dedicated customer-oci.com path-style endpoints.
API example
Set provider to oci_storage and provide a supported OCI hostname. The following rule uses a virtual-hosted endpoint:
Certificate Transparency Monitoring is now generally available ↗ across all Cloudflare plans.
Alerts for certificates Cloudflare issues on your behalf (Universal SSL renewals, backup certificates, Advanced Certificate Manager, Total TLS) are now automatically filtered out. Alert emails are also clearer and more actionable, with structured certificate details and a direct link to manage CT Monitoring in the Cloudflare dashboard.
Cloudflare Email security now lets administrators write their own content-based blocking rules. A new Blocked content area under Policies & rules lets you define a plaintext string or a regular expression, choose whether to scan the message subject, body, or both, and automatically block any message that matches.
Create rules using either plaintext matches or regular expressions — useful for blocking targeted phishing campaigns, known-bad phrases, or content patterns unique to your organization.
Choose the search location for each rule: subject, body, or subject and body.
Use the built-in regular expression checker to validate your pattern against sample text before saving, so you can confirm the rule matches what you expect and avoid false positives.
Matching messages are marked with a malicious disposition and prevented from reaching users' inboxes.
Blocked content rules currently only support the block action.
This feature is available for the following Email security packages:
Infrastructure applications support independent multi-factor authentication (MFA) with FIDO2 keys. You can allow ssh_fido2_key, piv_key, or both in application-level and policy-level MFA settings.
Users enroll FIDO2 keys through the App Launcher and connect with the generated SSH identity. FIDO2 keys for SSH are separate from browser-based WebAuthn security keys and Personal Identity Verification (PIV) keys.
Cloudflare Gateway now automatically detects Model Context Protocol (MCP) ↗ traffic flowing through your network. MCP is the standard protocol used by AI agents to connect to external tools and data sources. Gateway identifies MCP requests by inspecting protocol-specific headers and payload characteristics.
MCP policy selector
A new Is MCP selector (experimental.is_mcp) is available in HTTP policies. Use this selector to build Gateway rules that allow, block, or isolate MCP traffic.
This selector is currently in beta and may change before general availability.
For example, the following policy blocks MCP traffic that does not arrive through an approved MCP portal:
Selector
Operator
Value
Logic
Action
Is MCP
is
True
And
Block
Traffic Source
is not
MCP portal
AI security report
A new AI security report dashboard under Insights & Logs > Dashboards provides visibility into MCP usage across your organization. The dashboard includes:
Total MCP request volume, unique users, and unique MCP servers
A timeseries chart of unique MCP servers observed over time
A summary of Gateway policies that target MCP traffic