Hot take that isn't really a hot take:
The security community is quietly outsourcing the exact cognitive capability that makes it valuable.
Braindumped how to address it:
👉
Posts by Kreep
Defcon 33 is just around the corner (August 7-10, 2025, Las Vegas)! Hacking season is upon us.
I've put together some tips for #DEFCON first-timers on navigating the conference and Vegas itself. Hoping it helps peeps make the most of the experience!
kreep.in/desert-ops-v...
Just dropped my BadSuccessor .NET PoC showing how to abuse Delegated MSAs for AD access:
✔️ OU discovery
✔️ Zero-creds MSA creation
✔️ Works with user or machine account
Details + code:
🔗 github.com/ibaiC/BadSuc...
🧵 kreep.in/badsuccessor...
Been researching how to silently freeze Windows processes to hijack them during red team ops. Target apps like Teams, Slack, Outlook—pause them without breaking UI. Useful for stealthy social engineering.
Write-up here:
🔗 kreep.in/friendlyfire...
BOF:
💻 github.com/ibaiC/Friend...
Microsoft is moving key security mechanisms back to userland. On the surface, this should make life easier for threat actors—but surely they’ve thought of that, right? 🤔
At least we're getting more control over our machines again.
A nice reminder that everyone underrates their skills and talent www.youtube.com/watch?v=dZCr...
The #CrowdStrike2025 report is wild! 🚨 Breakout times as low as 51 sec, a 442% surge in vishing, and attackers using #GenAI for social engineering. Things are changing, FAST.
#CyberSecurity #Infosec
I've been diving into Windows Security Internals - James Forshaw & Evading EDR - Matt Hand.
They've been easy to get through and definitely filling some knowledge gaps.
Any other must-reads in the Windows internals or RT realm? 👀 📖
AI vs AI. We're really doing this.
RedAgent is an LLM designed to jailbreak chatbots for pentesting & red teaming. So now we’ve got AI breaking AI, patching itself, and breaking again.
Are we advancing tech or just setting up a whole new battlefield?
🔗 https://arxiv.org/abs/2407.16667
Screenshot showing the execution of the proof-of-concept named PowerChell in comparison to a typical PowerShell prompt. In particular, it shows that PowerChell is able to bypass the Constrained Language Mode (CLM).
In this blog post, I explain how I was able to create a PowerShell console in C/C++, and disable all its security features (AMSI, logging, transcription, execution policy, CLM) in doing so. 💪
👉 blog.scrt.ch/2025/02/18/r...