Yeah... mine got accepted, but there is no way to actually activate it as per the instructions ( the page is missing) and the customer support bot immediately drops the session when you mention CVP so I dunno how to feel.
On the other hand GLM is quite nice :)
SkelSec retweeted
It’s fun watching companies build paid pentest automation on top of open source tools and research, and give nothing back to the community they’re built on.
Historically, this space has been pretty good about crediting the original researchers and contributing back. That seems to be slipping as everyone tries to productize it.
The pullback in community contributions over the last couple of years is pretty obvious. I’m not sure many people realize what their R&D costs will look like if that continues, even with AI.
New aardwolf version is out! github.com/skelsec/aardwolf/…
Contains fixes (thanks to contributors) and new features like full compression suite support in Rust, clipboard file upload/download, drive redirection MS-RDPEFS as well as extensive test cases
Also tested by @al3x_n3ff
SkelSec retweeted
❗New research: Timo Longin (@timolongin) found two ways to spoof arbitrary icloud.com senders in @Apple's mail infrastructure - passing SPF, DKIM & DMARC.
Fixed after coordinated disclosure, $15,000 bounty paid👉 r.sec-consult.com/icloud
#EmailSecurity #SMTPSmuggling
I believe @MiaAI_lab is personally responsible for a large fraction of spark sales globally.
Thank you for your work!
Qwen3.8-Flash-Next for a single DGX Spark got a serious upgrade with TensorFold🔥
This is a completely new recipe, optimized and tuned for TesnorFold! Expect further improvements!
- KV cache pool is ~1.3M
- Default context 256k, with 5 concurrent.
- Faster everything compared to vLLM!
Performance:
Decode prose 62+ tok/s single stream
Decode prose 119+ tok/s on 5 streams
Prefill is mostly 2500 tok/s across the board!
In addition, expect TensorFold recipes for GLM 5.3 Flash and DeepSeek v4.1 Flash - coming soon!
Get it here:
github.com/MiaAI-Lab/Qwen3.8…
SkelSec retweeted
Qwen3.8-Flash-Next for a single DGX Spark got a serious upgrade with TensorFold🔥
This is a completely new recipe, optimized and tuned for TesnorFold! Expect further improvements!
- KV cache pool is ~1.3M
- Default context 256k, with 5 concurrent.
- Faster everything compared to vLLM!
Performance:
Decode prose 62+ tok/s single stream
Decode prose 119+ tok/s on 5 streams
Prefill is mostly 2500 tok/s across the board!
In addition, expect TensorFold recipes for GLM 5.3 Flash and DeepSeek v4.1 Flash - coming soon!
Get it here:
github.com/MiaAI-Lab/Qwen3.8…
SkelSec retweeted
Full disclosure - iDiskp64, a BYOVD-capable driver with an unusual activation gate - can be satisfied with a programmatically crafted virtual disk file (poc provided). github.com/Atos-TRC/full_dis…
SkelSec retweeted
As Flare-On starts I am happy to announce the official Hex-Rays IDA MCP Server is out! 🥳
Details and links below ⬇️

ALT We are happy to introduce the official IDA MCP server! This free and open source software connects your IDA installation with AI agents, enabling them to disassemble, decompile, and support reverse engineering tasks. It works great with models like Gemini, Qwen, or Opus using familiar harnesses like Claude Code, Codex, or Pi. Across our internal malware analysis-focused benchmark, IDA MCP’s “code mode” architecture reduced token consumption by approximately 20% compared to popular alternatives.
SkelSec retweeted
🔓 Your first hardware hack might end with a BitLocker key.
For many students, Breaking BitLocker is their first time:
→ opening a laptop to find attack points
→ microsoldering onto a PCB
→ connecting a logic analyzer
→ watching TPM traffic on a real hardware bus
→ extracting the key that actually unlocks the disk
That moment when random electrical signals suddenly become understandable data — and then a BitLocker key — never gets old. 🔬⚡
This December we’re bringing the latest Breaking BitLocker to Black Hat Europe 2026 in London 🇬🇧
And we’ve made the hardware side even more accessible.
We built our own LPC + SPI decoders for PulseView, so instead of treating the logic analyzer as a magic box, you can actually SEE the communication, understand it and watch the attack unfold on the bus.
Over two very hands-on days:
⚡ TPM Bus Sniffing
⚡ LPC/SPI logic analysis
⚡ BitLocker key extraction
⚡ DMA attacks
⚡ Bootloader-based attacks
⚡ Current BitLocker research & mitigations
⚡ Most importantly: understanding why an attack works - and when it doesn’t
No previous hardware-hacking experience required.
🧰 All attack hardware + adapters included
💻 Prepared target laptop included
🔬 You perform the attacks yourself
🎒 And the equipment is yours to keep afterwards
One important thing: we still need additional Early Bird registrations for the Black Hat class to go ahead.
So if Breaking BitLocker has been sitting on your training wish list, don’t wait for regular registration - there may not be one.
👉 hos.direct/bheu26
Know a forensic practitioner, pentester, security researcher or LEA colleague who should finally cross the software/hardware boundary?
Send this their way. Let’s get another group around the workbench at #BHEU. 🔓🔬
#BlackHat #BitLocker #HardwareSecurity #DFIR #TPM #InfoSec
SkelSec retweeted
If you're interested in SDR / RF security and looking to attend @BlackHatEvents EU, make sure to get your early bird tickets with discount pricing before September 25th!
👉 Black Hat EU (London), December 7-8 2026: blackhat.com/europe/training…
#SIGINT #RFsecurity #SDR
SkelSec retweeted
A Tale of Several Hijacks and What It Taught Me About Runtime-Driven Testing - if you fancy reading a story about simple vulnerabilities (Ghostscript CVE-2026-19547 and similar in Apache, PHP, GnuPG, wget, Microsoft Coreutils) and with an interesting moral, here's my latest article 😉
atos.net/en/lp/cybershield/a…
Meet Octoagent, Octopwn's on-prem, model agnostic internal network pentesting agent, boosted by local or cloud LLMs.
Fully onprem, controlled, customizable, transparent and logged.
youtu.be/f_HqEnIe1zM
SkelSec retweeted
New gowitness, 3.20! github.com/sensepost/gowitne…
Friday afternoon (vibe)coding project that was on my to-do for a while: obtaining Entra ID tokens from an endpoint by asking the WAM. This alternative to using the PRT cookie follows the legitimate SSO flow used by apps like Teams to obtain tokens. Code: github.com/dirkjanm/askWAM

























