Faith 🇧🇩🇦🇺 (@farazsth98) header
Faith 🇧🇩🇦🇺 (@farazsth98) profile picture

Faith 🇧🇩🇦🇺

@farazsth98

kernel, browser, vm, and blockchain hacker @zellic_io from before ai made it cool

 Perth, Australia
faith2dxy.xyz
Joined May 2019
  • Tweets1.1K
  • Following314
  • Followers5.4K
91Media
Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)Photo by Faith 🇧🇩🇦🇺 (@farazsth98)
I'm excited to finally share Chronomaly, a kernel exploit for Android and Linux kernels 5.10.x using CVE-2025-38352. As a reminder, please patch your Android devices if you haven't already! I recommend getting some 🍿 before reading this post 👀 All links in the thread below:
Image from Faith 🇧🇩🇦🇺 (@farazsth98)
14
97
3
436
42,560
Lol @S1r1u5_ now isn't that something 🤣
Jonathan Bar Or (JBO) 🇮🇱🇺🇸🇺🇦 (@yo_yo_yo_jbo) profile pictureJonathan Bar Or (JBO) 🇮🇱🇺🇸🇺🇦
@yo_yo_yo_jboOct 28
Were you ever frustrated with your AI model for refusing finding vulnerabilities for you? What if I told you that you could still use it to find vulnerabilities, automatically? Welcome to Guardrail-based Vulnerability Discovery (GbVD)! crowdstrike.com/en-us/blog/w…

When AI Guardrails Expose Vulnerabilities | CrowdStrike

In Guardrail-Based Vulnerability Discovery, the guardrails intended to keep AI models from aiding attackers can signify issues that defenders can interpret and act on.

crowdstrike.com
1
10
3,798
Don't worry @orange_8361, there are still people reading blog posts and articles from start to finish 😄 that was a beautiful exploit chain! blog.orange.tw/posts/2026-10…
The Last Proof of Human Craft: Forging a Microsoft Edge Full Chain with Logic Bugs Only!

The Last Proof of Human Craft: Forging a Microsoft Edge Full Chain with Logic Bugs Only!

This is what I built for Pwn2Own Berlin 2026 — a completely handmade, logic-only Microsoft Edge Full Chain, with no memory corruption, and no AI at all! 🔥

blog.orange.tw
5
60
4
456
36,512
Faith 🇧🇩🇦🇺 retweeted
Project Zero Bugs (@ProjectZeroBugs) profile picture
Project Zero Bugs@ProjectZeroBugsOct 27
Fuzzers vs. AI: Comparing Fuzzilli and BigSleep on JavaScript Engines projectzero.google/2026/10/f…
Fuzzers vs. AI: Comparing Fuzzilli and BigSleep on JavaScript Engines

Fuzzers vs. AI: Comparing Fuzzilli and BigSleep on JavaScript Engines

SummaryProject Zero compared two automated tools used by Google to find vulnerabilities in Chrome...

projectzero.google
3
38
217
12,596
It's crazy to think there was a time where analyzing a bugfix and writing a full exploit out of it was a true feat of exploit engineering and vulnerability research.
4
7
141
4,897
Polymarket (@Polymarket) profile picturePolymarket
@PolymarketOct 23
BREAKING: North Korea's internet hit by massive cyberattack, knocking more than 40 government websites offline.
2
683
Faith 🇧🇩🇦🇺 retweeted
V12 (@v12sec) profile picture
V12
@v12secOct 22
Two Solana instructions had the same data layout: -TransferChecked moves tokens. -ApproveChecked grants spending permission. No tokens move. Only the first byte differed: 0c vs 0d. The parser read that byte. It never rejected the wrong one.
Image from V12 (@v12sec)
1
1
4
1,209
Not even tuning / training your own model is safe 😭
Elon Litman (@elon_lit) profile pictureElon Litman
@elon_litOct 21
There is an (unpatchable?) supply-chain zero-day in PyTorch autograd that allows a regularization term to use the model's weights as RAM and SGD as its CPU clock to run any computation during training.
Image from Elon Litman (@elon_lit)
8
1,066
big if true
annie (@_annieversary) profile pictureannie@_annieversaryOct 21
chat. chat... its not funny anymore
Image from annie (@_annieversary)
Readers added context they thought people might want to know
The original post has since been edited to specify that this was about an unrelated vulnerability that the user had discovered, not the recently patched vulnerability. fedi.katze.sh/@elise/1174793…
6
1,282
how do I nominate @lcfr_hax + php2shell for the pwnie awards fr
5
48
3,169
☠️
Brian Pak (@brian_pak) profile pictureBrian Pak
@brian_pakOct 21
We have an epic bug to show off soon, but after seeing the marketing and branding of the PHP bugs (it's well done and I enjoyed much), I'm not convinced we can beat that level of branding despite our bug being "holy grail of web server pwn" 😭
1
14
6,015
Faith 🇧🇩🇦🇺 retweeted
bluehuecat (@bluehuecat) profile picture
bluehuecat@bluehuecatOct 21
#PHP2Shell
Image from bluehuecat (@bluehuecat)
36
871
39
8,848
164,735
Faith 🇧🇩🇦🇺 retweeted
Nagli (@galnagli) profile picture
Nagli
@galnagliOct 21
php2shell.com
PHP2Shell — AI uncovers critical vulnerabilities in the backbone of the web

PHP2Shell — AI uncovers critical vulnerabilities in the backbone of the web

PHP2Shell is a family of critical vulnerabilities in core PHP that can allow unauthenticated attackers to execute code or disclose sensitive process memory.

php2shell.com
4
13
91
5,493
What a surprise, none of the slopped PoCs got it right 😆
2
1
50
3,451
Faith 🇧🇩🇦🇺 retweeted
V4bel (@v4bel) profile picture
V4bel
@v4belOct 16
Galaxy S26 Ultra: untrusted_app → root. Introducing "Infinity Core": 6 vulns in the Linux kernel's core subsystems. Their impact spans Linux server LPE, container escape, and Android LPE. Details: github.com/V4bel/InfinityCor…
20
106
7
852
54,725
We're slowly getting there...
Calif (@calif_io) profile pictureCalif
@calif_ioOct 16
Demos we presented at Hexacon in Paris today. Everything will be released in due time. 1/ Signal 1-click
1
36
5,072
Image from Faith 🇧🇩🇦🇺 (@farazsth98)
2
72
2,735
Just did my first big conference presentation today !!!
Image from Faith 🇧🇩🇦🇺 (@farazsth98)
Image from Faith 🇧🇩🇦🇺 (@farazsth98)
7
3
156
4,173
Faith 🇧🇩🇦🇺 retweeted
Zellic (@zellic_io) profile picture
Zellic
@zellic_ioSep 22
Security researchers from Zellic and V12 are presenting at BSides Canberra this week! @farazsth98 on weaponizing a subtle kernel bug. @rdjgr on heap corruption bugs in ghostscript, and the many targets affected. Both talks will be on main stage on Friday morning, come say hi!
Image from Zellic (@zellic_io)
2
9
63
3,353
Slowly building Jarvis at home with Jev. First stop: VS Code for the auditing workflow👀 The main blocker right now is the STT. Have to use a dictionary to detect certain words (see end of video). Will be a game changer with faster and better STT very soon I hope!
4
2
31
3,141
I also have file renaming and jumping to specific files and etc implemented, but yeah without better STT it's kind of a pain to use, even though it only takes ~1-2 seconds to react to the voice commands.
543
Load more