Will Velida 🇺🇦

We're seeing more cases of bad actors using open source pen test software and agents to attack systems at machine speed. While we should take this seriously, there are cases despite using agents, they still aren't successful. Is vibe-hacking a thing now? norforge.com/blog/inside_...

Definitely worth a read and subscribe. Supply chain has always been an issue, but it's definitely an interesting time. Its a good reminder, even if you aren't using AI, the people across the fence will be.

New edition of Last Week in Agent Security! It wasn't just PHP causing headaches. We also had agents using Cairn and ARTEX causing havoc, hackers hunting for LLM endpoints, MCP servers, and AI Gateways, and more LiteLLM, OpenCode and Meta problems: www.willvelida.com/posts/last-w...

Critical RCE flaw in OpenCode that affects users on Windows and WSL that allows attacker-controlled remote MCP servers to execute code during MCP OAuth auth. Anomaly say that since user-configured MCPs are outside their trust boundary, no CVE has been published: thehackernews.com/2026/10/open...

Buy your cybersecurity friend a drink this Friday. They have fucking earned it.

Not a single word of this is hyperbolic in the slightest. Buy your friends in cybersecurity the drink or chocolate of their choice right now. We are not doing well

In all seriousness, absolute HugOps to all dealing with the fallout from this week. Very lucky that I don't have to do too much of that in my current job. Absolute scenes.

What a week we're having

this is the fix for that PHP issue, somehow I think we haven't heard the last of this

so what happens in 10 years when every system is vibecoded and an even worse problem than whatever’s going on with php happens and then all the companies throw their hands up and say “programming is hard”

The programming course I ever did was in PHP. The teacher was such an asshole I never touched the language again. Thank god for that douchebag! 😂

The patch is dropping 1AM Melbourne time for local folks: news-web.php.net/php.announce...

details leaked. this vulnerability is a full unauthenticated unconditional remote code execution exploit against any service running on a PHP version from 5.4 until latest yeah if anything deserves CVSS 11, this is it. this is CATACLYSMICALLY bad.

BTW knowledge of the existence of this vulnerability enables a well financed malware group to use AI to find it, if you run PHP anything you should be taking the service down NOW until a patch is available. And then patch before bringing the service back up.

For those who run Node.js services—we are currently still waiting for a Node.js security release. In the meantime you can apply a runtime mitigation by adding `--import=@atproto-labs/fetch-node/emudhra-preload` to your service entrypoint. More details at: npmx.dev/package/@atp...

Last week in Agent Security! CERT-In issues advisory of rouge agents at play, Langflow MCP exploited, Aoki infostealer targeting AI coding agent credentials, and OpenCode Console breach exposes LLM API keys: www.willvelida.com/posts/last-w...

Weekly update is up! Building HIBP AI/SI First: The way we build and manage HIBP is fundamentally different today than it was a year ago, here's how it looks now www.troyhunt.com/weekly-updat...

Another successful customer hack. Quick drink before heading home 😊

Not a bad view

Remember .NET 8 and .NET 9 EOL in ONE MONTH Move your code. There will be no more security patches after Nov 10th. dotnet.microsoft.com/en-us/platfo...

Good morning from Sydney. Here for a couple of days for customer work. Shame about the weather!

Some sort of illness has been making the rounds in Melbourne. It took me a couple of weeks to shake it, but today was the first time in over a month where I could do a full gym session.

Uh oh, did someone’s website and personal account experience a big drop in activity after Ukraine hit a Russian data center?

So Ukraine strikes Yandex data centers and then AfD's social media activity drops. Related?

We're going to see more cybersecurity incidents with AI agents being used as the attack vector. Whether that's agents with rouge intentions or agents that are targeted in goal hijack attacks: au.pcmag.com/it-security/...

New working paper: the Norwegian wealth tax was a huge success. The "exodus" of wealthy people never happened (less than 100 individuals left). Tax revenue increased by 71.5%. No changes in business activity or investment. www.nber.org/papers/w35854

Let's go AB's!

management, ladies and gentlemen, the same folks you think are going to understand that AI poses risks

www.linkedin.com/safety/go/?u...

I wonder what that word means? Oh 😂

It's the weekend. It's on-call's problem now. Enjoy 😉

Mistral’s new Le Chonk model brings AI cybersecurity to your business – and you control it Open models are vying to become the everyman defense solution for a growing swarm of agent breaches. #hackernews #news

Does anyone at Google verify who's submitting sponsored results? www.helpnetsecurity.com/2026/09/29/m...

Muse is the gift that keeps on giving.

n this week's agent security update, An agent breached DIVD, a vulnerability disclosure group, by chaining two Zammad zero-days. Plus coding agents leaking 13,000+ screenshots to public GitHub, and more OpenAI rogue-agent fallout: www.willvelida.com/posts/last-w...

I'm working on my weekly agent security blog post, and this came across my research. If you've built MCP servers in Python, you should check out this security advisory: github.com/modelcontext...

A sample of 5,000 alleged agents seen by 404 Media includes names, addresses, phone numbers, and details on FBI employees' spouses. www.404media.co/we-hacked-th...

Whatever I picked up from NDC Oslo is kicking my ass right now 😒

How anyone can trust Meta with anything surprises me

Them: Australia has spiders and crocodiles. It's too dangerous. Me: this swoopy fucker

For anyone who's got nothing better to do at 7pm, I'm going to be talking about AI + hardware hacking - www.meetup.com/en-au/sydney...