AI
How we caught the Axios supply chain attack--with a proof of concept tool built in an afternoon
Last Monday night I was working late and a Slack alert came in from a monitoring tool I had built three days earlier. Axios compromised; one of the most popular npm packages in the world.
The idea: monitor changes as they get pushed to package repos. Run a diff to see what changed. Use AI/LLM to determine if the changes are malicious. That's basically it--Supply Chain Monitor.
I never got a single false positive.
The BuddyBoss Attack: Claude’s Supply-Chain Attack
Researchers discovered the attacker’s infrastructure on 18th March 2026 and obtained the complete Claude chat logs used to develop and execute the attack chain.
The attacker was low-skilled; and told Claude they were doing a CTF. Claude did almost all the work, developing malware, finding a way around Cloudflare, uploading it, testing it to make sure it entered the supply chain, and gathering the responses from the victims.
Claude Code cracks FreeBSD within four hours
He identified a vulnerability in the FreeBSD operating system and exploited it within four hours. Claude was also capable of creating a working exploit. The vulnerability has been reported as CVE-2026-4747.
Politics
'What a revolting message' - Amnesty International head on Trump's threat
"Iranian civilians will be the first to suffer from the destruction of power plants and bridges".
"No more electricity, heating, or water; unable to flee the attacks. Potential cascading war crimes."
It comes days after a group of more than 100 experts on international law signed an open letter expressing "profound concern" about what they see as serious violations of international law by the US, Israel and Iran in the war
Trump’s new app has no privacy policy and uses Russian software
Donald Trump has been promoting the White House’s new mobile app — pushing it to become the third-most downloaded item on Apple’s popular App Store. But the app reportedly has numerous cybersecurity vulnerabilities, does not properly disclose the data it shares, and uses software components from a Russia-founded company.
"It shares users’ IP addresses, time zones and other data to third-party services,” NOTUS reports. “But most of its users wouldn’t know that, because the app doesn’t disclose its data sharing the way most others do.”
CBP facility codes sure seem to have leaked via online flashcards
It exposed highly confidential information about security procedures in US Customs and Border Protection facilities around Kingsville, Texas, such as checkpoint door codes.
Commanders now responsible for cybersecurity training after Army cuts online course requirement to once every 5 years
Meta Pauses Work With Mercor After Data Breach Puts AI Industry Secrets at Risk
Mercor is one of a few firms that OpenAI, Anthropic, and other AI labs rely on to generate training data for their models. The company hires massive networks of human contractors to generate bespoke, proprietary datasets for these labs, which are typically kept highly secret.
SandyClaw, a tool that scans OpenClaw skills and assesses their safety
Very early product, not open to the public yet.
Top FEMA Official Doubles Down on Claim He Teleported to Waffle House
Infosec
New Rowhammer attacks give complete control of machines running Nvidia GPUs
GDDRHammer, GeForge and GPUBreach hammer GPU memory in ways that hijack the CPU.
“By corrupting GPU page tables, an unprivileged CUDA kernel can gain arbitrary GPU memory read/write, and then chain that capability into CPU-side escalation by exploiting newly discovered memory-safety bugs in the NVIDIA driver,” the researchers explained. “The result is system-wide compromise up to a root shell."
TasksJacker: Latest DPRK Attack Skips the Fake Interview and Goes Straight to Compromising GitHub Users
What makes TasksJacker novel isn't just its scale—it's the attack vector. By weaponizing VS Code's tasks.json auto-execution feature, attackers have created a scenario where simply opening a cloned repository in your IDE can compromise your system. No user interaction required beyond a git clone and opening the folder.
Device code phishing attacks surge 37x as new kits spread online
Device code phishing attacks that abuse the OAuth 2.0 Device Authorization Grant flow to hijack accounts have surged more than 37 times this year.
In this type of attack, the threat actor sends a device authorization request to a service provider and receives a code, which is sent to the victim under various pretexts.
Next, the victim is tricked into entering the code on the legitimate login page, thus authorizing the attacker's device to access the account through valid access and refresh tokens.
This flow was designed to simplify connecting devices that do not have accessible input options (e.g., IoT devices, printers, streaming devices, and smart TVs).
The CertGraveyard
A repository of abused code-signing certificates, so they can be revoked by CAs.
Cisco Patches 9.8 CVSS IMC and SSM Flaws Allowing Remote System Compromise
Linux Foundation is Launching the x402 Foundation and Welcoming the Contribution of the x402 Protocol
The x402 protocol from Coinbase is a universal standard for payments that embeds payments directly into web interactions, enabling AI agents, APIs, and apps to transact value as seamlessly as they exchange data.
A Secure Chat App’s Encryption Is So Bad It Is ‘Meaningless’
TeleGuard is an app downloaded more a million times that markets itself as a secure way to chat. The app uploads users’ private keys to the company’s server, and makes decryption of messages trivial. An attacker can trivially access a user’s private key and decrypt their messages. It’s possible to retrieve a specific user’s private key by simply plugging their user ID into TeleGuard’s API.
|