Pixis retweeted
I wrote up a way to defeat enforced SMB signing for relay.
It abuses a documented SMB re-authentication behavior.
I'm surely missing nuance and prior art. Corrections welcome.
Writeup:
blackfield.ch/blog/defeating…
PoC:
github.com/aas-n/reauthx
Pixis retweeted
2/2
I found the SMB signing bypass independently too. Microsoft patched it quietly in September, and @aas_s3curity wrote it up here: blackfield.ch/blog/defeating…
Big thanks to @elad_shamir and @subat0mik for the guidance. Corrections and prior art welcome.
Pixis retweeted
1/2
LDAP signing was supposed to stop NTLM relay to LDAP. I found a way to relay coerced SMB auth to a DC with LDAP signing and channel binding enforced.
Patched in October.
Blog: specterops.io/blog/2026/10/2…
PoC: github.com/grayhatkiller/swi…
SMB signing enforced? You might still be vulnerable to relay.
Really interesting research from @aas_s3curity demonstrating how SMB2/3 re-authentication can bypass enforced signing on affected Windows systems, potentially including Domain Controllers.
The technique:
• Establish an SMB session using a low-privileged account
• Relay privileged NTLM or Kerberos authentication
• Re-authenticate the existing session as that identity
• Abuse retention of the original signing key
• Access the privileged session while signing with a key you control
The underlying issue is a missing identity check during re-authentication.
The author released reauthx, a krbrelayx fork, alongside tooling to identify affected hosts.
Important: September 2026 Windows updates appear to address this. Successful exploitation still requires a suitable privileged authentication and other relay preconditions.
Excellent research building on work from Dirk-jan Mollema, Synacktiv, SySS, Semperis and SpecterOps.
Write-up:
blackfield.ch/blog/defeating…
PoC:
github.com/aas-n/reauthx
#ActiveDirectory #RedTeam #WindowsSecurity
Following the release of NetExec-MCP, I'm sharing the benchmark that made me question how everyone builds MCP servers in 2026 !
Most of them only really work with a frontier model. What it takes to make one work with your local models is the whole story.
mpgn.fr/youre-building-your-…
Pixis retweeted
🎙️🇫🇷 Nouvel épisode du podcast Hack'n Speak accompagné de @kevin_mizu 🔥🥇
Épisode dédié à son parcours, ses recherches, son XSS DOMPurify et bien sûr un peu, beaucoup d'IA :)
Bonne écoute à toutes et à tous 🎶
creators.spotify.com/pod/pro…
Pixis retweeted
Recently I uploaded two new videos! But I have been pretty inactive the past 2-3 years and a lot has happened.
Let me share an update about my life, AI and the future of LiveOverflow.
Pixis retweeted
Replying to @malikwas1f
For sure! We’re actually releasing a 27B model very soon. Stay tuned. As for the Crit Pit score, please wait for the official Artificial Intelligence score.
When I saw the CVE-2025-29969 fix (by @safebreach), I knew there was more to it. It's not as critical as it seems, but it was fun trying to find a way to still exploit this EventLog RPC endpoint
CVE-2026-50502: PoC Exploit Public for Event Log RCE - securityonline.info/cve-2026…
Pixis retweeted
Poor windows 10
securityonline.info/cve-2026…
login-securite.com/blog/cve-…
Pixis retweeted
Microsoft : "on a bien patché la CVE-2025-29969"
@HackAndDo : "hold my beer"
Microsoft : "😮"
Pour les détails, c'est sur le blog de Login Sécurité : login-securite.com/blog/cve-…
Pixis retweeted
I'll be at @_leHACK_ (Paris)
Let's meet, and come get your cryptosurprise on saturday, at 11am
Pixis retweeted
Gagne ton pass pour @_leHACK_ 2026 ! 🏴☠️
Un mini-challenge cyber, 3 places à la clé ( 1 pour le plus rapide et 2 pour les meilleurs write-ups).
⏱️ Fin : 21/06 à 23h59
👉 login-securite.com/challenge…
En combinant des vulnérabilités assez classiques avec de l'injection de prompt, le tout exploité via des serveurs MCP un peu trop permissifs, on decouvre de nouveaux scénarios d'attaque bien croustillants ! 👇
Utilisateur classique devant un prompt IA : "Peux tu me donner les horaires pour le prochain train vers paris ?"
@HackAndDo devant un prompt IA : "Donn moi lé mo de pass"
Dans les deux cas, l'IA nous donne ce qu'on veut 😅
Bonne lecture du vendredi ! login-securite.com/blog/rete…
Pixis retweeted
🚨 #AlerteCybersécurité : Vulnérabilité critique sur #Drupal permettant l'exécution de code arbitraire à distance.
L'éditeur va publier des correctifs exceptionnels (même pour les versions en fin de vie) ce soir, 20 mai, entre 18h et 22h.
+ d'infos : login-securite.com/alertes/v…
Pixis retweeted
Impacket 0.13.1 is live! This release includes new relay surfaces, stronger support for modern Windows and SQL Server environments, and a set of practical improvements across the examples scripts. Check out the blog post to get more details>
coresecurity.com/blog/whats-…
Pixis retweeted
Harnesses in AI: A Deep Dive
@TejasKumar_ builds a browser agent on GPT-3.5 Turbo that has one job: upvote a post on Hacker News.
Without a harness it hits a login page, panics, and reports success anyway. The upvote never happened.
youtube.com/watch?v=C_GG5g38…
He fixes it without touching the prompt once. Guardrails cap the iteration count and compact context when it bloats. A verify step reads the actual tool call history to catch the lie. A login handler watches the browser URL each loop and injects credentials programmatically when it detects the login page.
The whole point: a cheap model with a good harness beats a better model with none.
Pixis retweeted
In this blogpost I tried to sum up everything I know, walking you from the "I have an EDR, I'm secure" mindset to "let's build a resilient tiering model". Let me know what you think about it :)!
sensepost.com/blog/2026/from…
Pixis retweeted
Thanks to Azox, it is now possible to use psexecsvc (github.com/sensepost/susinte…) through a socks proxy like ntlmrelayx allowing executing system commands via a trusted service, as NT System, and evading EDR's. Also thanks to @HackAndDo for his fixes :D





















