Localizing compliance at scale: achieving South Korea’s ISMS certification
For ten days, auditors appointed by the Korea Internet & Security Agency (KISA) sat together with us and worked through 80 controls. That is what it took for a company headquartered in the United States, operating a global network serving Korean users, to be certified in South Korea. Cloudflare has now obtained South Korea's Information Security Management System (ISMS) certification under the ISMS-P framework.
For customers in Korea, this is more than a compliance checkbox. It is evidence that Cloudflare's security and privacy program has been examined not only for whether our controls exist, but for how they operate in the context of Korean law, the Korean language, and the particular expectations of Korean regulators.
What a Korean user sees when they sign up
The most visible result of this work is the Cloudflare account sign-up page. In most parts of the world, the line that sits above the sign-up button reads: "By continuing, I agree to Cloudflare's terms, privacy policy, and cookie policy."
A user signing up from South Korea sees something different. The Korean experience is triggered by the country selected at sign-up and by IP geolocation from within South Korea, and it breaks the same Continue reading
Using External Tools in BGP Labs
After Kazuya Takahashi released his wonderful network-xray tool and added it as an external tool to the open-source BGP configuration labs, I wrote a short document explaining how to start external tools in your BGP labs without changing the lab topology, and how it works in GitHub Codespaces (hint: wonderfully).
I also modified all lab instructions to include a pointer to the optional tool-setup step.
Happy labbing ;)
Attention people who think UB is fine
After my post on undefined behavior in C some people insisted that it’s demonstrably fine, we know what the code will do, and at worst we’ll get a segfault.
This is false.
Here’s an example.
C code
#include<utility>
// Let's pretend this is a memory mapped device controlling
// motors and gears and stuff.
volatile int device_command = 0;
void serve_hot_chocolate() {
device_command = 42;
// Oh no, we spilled some hot chocolate, invoking UB.
//
// The definition of `std::unreachable()` is:
// Invokes undefined behavior at a given point.
std::unreachable();
}
void explode() {
device_command = 3;
}
int main() {
serve_hot_chocolate();
}
Example assembly output
With GCC 16.2 with -O1.
"serve_hot_chocolate()":
mov DWORD PTR "device_command"[rip], 42
"explode()":
mov DWORD PTR "device_command"[rip], 3
ret
"main":
mov DWORD PTR "device_command"[rip], 42
"device_command":
.zero 4
Ruh roh. Exploding chocolate! The function explode() will be executed, even
though it was never called!
Another example: dereferencing a null pointer
Some people insisted that dereferencing a null pointer is fine. It’ll just try to access address zero! And these people supposedly know what happens when you try: if nothing is mapped there, you’ll get a segfault.
Again, not true.
Let’s say we change Continue reading
SONiC in a BGP EVPN Data Center: Errata, Updates and Notes
Errata
No corrections reported yet.
Updates
No updates available yet.
Notes
- 21 October 2026: First edition published on Leanpub.com
- 22 October 2026: Higher resolution file Uploaded to Leanpub
Introducing full transaction detection for Cloudflare WAF
Today, we are expanding our web application firewall (WAF) by adding two new capabilities: full transaction detections and response rules. Full transaction detections evaluate each HTTP request alongside its response to identify evidence of exploitation. It helps answer both "Does this request look malicious?" and "What happened when the application handled it?" Response rules can then block the origin’s response before it reaches the requester.
These are valuable protections because the latest frontier models are uncovering vulnerabilities buried deep in widely used software, giving defenders clarity on what to fix — and attackers new paths to exploit. The recent php2shell vulnerability shows the stakes: a flaw in a common software component can put applications across the Internet at risk. AI agents scanning for vulnerabilities can be extremely persistent, adapting payloads using feedback, and chaining vulnerabilities, credentials, and permissions. An application's response can reveal whether an attempt worked and give an attacker information that helps them go deeper.
Inspecting incoming requests remains essential for stopping recognizable attacks before they reach an application. But some threats become clear only when the request is considered alongside what the origin returns. A suspicious input followed by a response exposing sensitive information tells Continue reading
Cloudflare deploys protections against php2shell vulnerability
Today, Calif.io, Wiz, OpenAI and Quarkslab disclosed two vulnerabilities in PHP, tracked as CVE-2026-109618 and CVE-2026-109619 and known as php2shell. The vulnerabilities affect all web servers running PHP 5.4 or later and they may allow an attacker to cause denial of service, remote code execution, and data exposure when an application processes specially crafted HTTP traffic.
Cloudflare has been part of a coordinated effort to deploy mitigations to protect the Internet, working with PHP foundation, OpenAI and Wiz among others. Cloudflare has developed and deployed web application firewall (WAF) protections since early October 2026, which were deployed about 9 hours ahead of public disclosure. Starting on October 21 3:02 am UTC, Cloudflare customers using our free or paid WAF were protected by default against php2shell.
Cloudflare successfully deployed multiple mitigations for all its customers while the vulnerability was leaked before the official release date and we observed attack traffic targeting our customers. We observed how attackers were mutating the attack payload to try to bypass our protections by tracking the volume of matches and cross-referencing with the rule logic.
Protecting against php2shell required more than standard request-based rules. Request data alone does not always provide enough evidence to distinguish an Continue reading
How Commerce.com and Cloudflare are getting storefronts ready for AI agents
AI can write code and pass the bar exam, but it still struggles to buy a pair of shoes online.
That’s because websites were built for humans, not agents. Most agents that shop for someone today work from a site’s HTML — they look at the page, guess which button does what, click, and try again when they get it wrong. This process is slow, uses a lot of tokens, and something as seemingly innocuous as a theme change can break it.
Enter WebMCP: a browser standard from Google that gives AI agents structured tools on a page, like search_products. Instead of hunting for buttons, the agent calls the exact tool it needs directly, making shopping faster, cheaper, and far more reliable. Because WebMCP is still an experimental proposed browser standard in early preview, it’s currently available behind experimental Chrome flags (chrome://flags/#enable-webmcp-testing), in Chrome Origin Trials, in some agentic browsers, and across developer environments like Cloudflare Kitesurf.
In August, we launched edge implementation of WebMCP in beta. Turn it on, and Cloudflare injects a tiny script at the edge to make those tools available automatically, without site owners writing a single line of code.
Crappy Configuration Errors: Cisco Nexus OS Edition
TL&DR: Programmers writing Cisco Nexus OS code should have known better. They didn’t (care?).
One can understand that an operating system written in the late 1980s generates crappy suboptimal error messages, even though it was known even in those days how to do things right.
Is it too much to hope that the programmers writing newer operating systems (like Cisco Nexus OS) would be better? Obviously, it is 🤦♂️
8 eMudhra certificates are compromised. Here’s what you need to know
On October 19, 2026, at approximately 17:00 UTC, Cloudflare became aware that private keys associated with eight eMudhra intermediate certificates were compromised. These certificates were widely trusted by browsers, operating systems, and also within Cloudflare. Because the keys themselves are weak, an attacker can create a forged certificate for any domain and use it to impersonate servers to intercept or modify traffic. This is a serious security incident with implications across the Internet. For reference, here is a link to more information about how SSL certificates work and why they are so critical to the operation of the Internet.
We immediately prioritized patching critical services across Cloudflare to protect our customers ahead of the public disclosure of the vulnerability. Our security review has not identified active exploitation of this issue on Cloudflare services.
We have released updates for the Cloudflare One Clients, also known as WARP or the 1.1.1.1 app, and Cloudflare Tunnel, also known as cloudflared. We have also deployed mitigations across our CDN, Developer, and Zero Trust products. If you are using Cloudflare Zero Trust with Gateway and an up-to-date client (see below), you are automatically protected.
Our response
We prioritized updating services that use public certificates Continue reading
Introducing tsrs: A port of TypeScript to Rust
Cloudflare and VoidZero are now working with Max Schwenk of Owner.com on tsrs, a Rust port of the TypeScript compiler. We are doing this work in close collaboration with the TypeScript team, and incubating tsrs as part of the Oxc JavaScript compiler toolchain.
Our goal is to help developers spend less time waiting for their code to be checked and more time building. Making TypeScript faster and more memory-efficient also helps bring complete development environments to more places, including right in your browser.
To be clear, we are not forking TypeScript. We believe the TypeScript team should remain in full control of the language’s design and direction. They have maintained a long-term vision while guiding TypeScript’s adoption across the JavaScript community. We do not want to change that.
The TypeScript team can move quite fast on many things, but for some features they may spend years intentionally exploring options and making prototypes before picking the right implementation. A Rust port has to follow those decisions closely, or it will eventually fall apart and become incompatible. We have to preserve the TypeScript language strictly based on its original implementation – and the TypeScript team should remain solely responsible for Continue reading
Testing OpenEIGRP with netlab
When Donnie Savage announced the open beta of OpenEIGRP, I quickly asked him whether it would be possible to get an easy-to-use container with OpenEIGRP integrated with other FRR daemons (and its CLI).
Donnie delivered ;) – you can download a ready-to-use container and try its EIGRP features. Let’s see how you can make that process even simpler with netlab:
Inspecting RustRadio and GNU Radio streams dynamically
When building software defined radio flowgraphs it’s common to want to inspect the live data as it flows through various stages. In GNU Radio, and RustRadio, the handoff points are very clear. They are the streams between the blocks.
In GNU Radio you can stop the graph, click a bit in gnuradio-companion, and have a new time sink (or whatever you need) added, and then start the graph up again. But it makes for a lot of stopping and starting. There’s also the drawback that it assumes that the GNU Radio flowgraph is graphical in the first place, and not running headless on a raspberry pi somewhere.
And for large graphs, preemptively adding GUI elements for everything you may possibly want to read at runtime can clutter things up, and waste CPU processing and rendering data you’re not interested in.
Another use case is if you have a constantly running RF sniff process using an SDR, to get a second reader of the same data, to help develop the next version without shutting down the one already running.
So clearly, what you want is to be able to connect to the flowgraph over the network, and start streaming any and Continue reading
Cutting our inference bill by 68% by being obsessive about optimization
When we previously blogged about our AI code reviewer in April, an average review cost a palatable $1.19. By the end of August, that figure had shot up to about $2.46. We hadn't made any significant architectural changes: the coordinator was still handing work to the same set of specialist reviewers, and the same risk tiers decided how intensive of a review we gave.
Around mid-September we tasked our agents with working out where those tokens were going and cutting them — aggressively. With code quality at high effort, the average review now costs just $0.77, down 68% from the $2.43 it averaged over our two most expensive weeks before we started. Our main goal was to cut costs, but as part of this work we also managed to optimize the reviewer, so that the median review now takes just 1m 32s instead of the 4m 09s it took before.
About 40% of the savings came from changes and optimizations to the reviewer's code, and the rest from experimenting with which models we use and how much effort they put in.
How a review doubled in cost
The chart below shows the cost of an Continue reading
Rolling the Root Key – WHat Happened?e
The Root Key of the DNS rolled to a new key on the 11th October. This key roll is special as it sits at the apex of the DNS hierarchy, and DNSSEC-validating resolvers need to add this key to their local trusted key collection to continue to validate DNSSEC-signed responses. How well did we do in configuring our DNS resolvers to track this key roll?Meet Tigera at KubeCon + CloudNativeCon North America 2026
Ask a platform team how many AI agents are running in their clusters and you usually get a pause. Agents arrive through application teams, not platform teams. They call tools over MCP, invoke other agents and reach external model providers, and a Kubernetes service account tells you which pod made the call but not which agent and on whose behalf.
That gap is what we’re bringing to KubeCon + CloudNativeCon North America 2026, November 9–12 in Salt Lake City. Find Tigera at booth 150 for quick Lynx sessions on identifiying, governing and containing AI agents, and pick up a new t-shirt while you’re there. On November 9, before the show floor opens, join us for a workshop on securing AI agents on Kubernetes and a happy hour with the Lynx community.
November 9: Workshop and happy hour
Both events are at Le Méridien Salt Lake City Downtown (131 S 300 W), about a five-minute walk from the Salt Palace.
Discovering and Securing AI Agents on Kubernetes
2:00–7:00 PM MST · Triumph Salon · In-person or virtual
Peter Kelly, CTO at Tigera, walks through the path we’ve watched teams take with agents in production, in order. Discovery comes first, because agents Continue reading
Introducing Clairic, a new System One class of extraction models
Today, we're announcing a preview of a new class of System One-style models: extraction models. Our first one, Clairic, is available hosted on Workers AI as @cf/cloudflare/clairic.We intend to release the model open-weights soon. Clairic is an extraction model that is able to extract, compute, estimate, and forecast values from unstructured inputs — all at a significantly higher accuracy, faster latency, and cheaper cost than traditional LLMs.
Context
When System One models were announced by TypeSafe a month ago, it took the industry by storm. As referenced by TypeSafe in their blog and defined by Daniel Kahneman’s book Thinking, Fast and Slow, a System One model should be fast and precise, which contrasts a System Two model that is capable of deliberate reasoning (like LLMs). With this, TypeSafe introduced Jev — a model that is purpose-fit for a specific use case, and priced accessibly given its niche domain. It excelled where LLMs failed, helping users make decisions based on probabilities and well-typed outputs. The introduction of Jev prompted people to start thinking about a new class of models that were not LLMs and how to integrate them into their workflows.
We were inspired by the concept of System One models and Continue reading