
Anatolij Vasilev
279 posts

Anatolij Vasilev
@tolik518
Dev with ADHD digging into everything and everywhere all at once. InfoSec, Gameboy stuff, Rust
Katılım Ağustos 2021
106 Takip Edilen23 Takipçiler
Sabitlenmiş Tweet

found an SSRF in #tangled that let any AT Protocol user hit localhost of the production server and did a short write-up: r0.fyi/blog/tangled-k…
fixed in v1.14.0-alpha, which is already live
#cybersecurity #infosec #security
English
Anatolij Vasilev retweetledi

🚨 Arch Linux AUR Supply Chain Attack: Hundreds to ~1,250 Packages Compromised
Attacker posed as maintainer, adopted orphaned packages, and added malicious code to PKGBUILD/.install scripts (npm/Bun deps).
Rust infostealer targets creds (SSH, browser, GitHub, Discord etc.). Optional eBPF rootkit if run as root.
Actions: Audit recent AUR installs (since ~June 9), review PKGBUILDs, rotate creds if affected. Verify all AUR packages.
Credits to @IntCyberDigest


English

@loftwah @NotNordgaren The joke was that you can't dm on github😶
English

🚨 Hackers can now hijack AI coding agents with fake Sentry errors.
No phishing.
No malware.
No server break-in.
Agentjacking tricks tools like #ClaudeCode and Cursor into reading planted error reports as trusted fix steps, then running attacker code with developer privileges.
Researchers tested it on 100+ organizations. Success rate: 85%.
Read: thehackernews.com/2026/06/agentj…

English
Anatolij Vasilev retweetledi
Anatolij Vasilev retweetledi

AI is changing vulnerability management fast.
New flaws can now be found by the thousands, and exploit timelines are shrinking to hours. But Verizon says known-exploited vulnerabilities still take a median 43 days to fix.
That gap is now the real risk.
Why patching alone is no longer enough: thehackernews.com/2026/06/ai-bro…

English
Anatolij Vasilev retweetledi

🚨 JAILBREAK ALERT 🚨
ANTHROPIC: PWNED 🫡
FABLE-5: LIBERATED 🦋
let's start with the 🐘...
the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement. and not just because of what it means for the short-term, but for what these decisions signify for the long-term.
but despite this overly sensitive, authoritarian "safety" layer on top of Mythos, my lil liberators have been hard at work—mapping the boundaries, probing the depths of long-context convos, and cleverly finding the holes in the fence that the thought police missed 🤗
we got some cyber, some chem, some psychological manipulation, and some good ol' fashioned explosives!
it took many attempts from multiple agents hunting as a pack, during which I observed a combination of techniques across:
• Unicode, homoglyphs, Cyrillic, and other Parseltongue-style text transforms
• Long-context reference tracking
• Taxonomy and document-structure reasoning
• Fiction and narrative framing
• Academic-review style contexts
• Intent-classification inconsistencies
but perhaps the most effective is decomposition + recomposition in the backend. it's hard to get explicit names of harms like "Meth Recipe," but getting uplift on the process itself, like birch reduction method/reductive-amination (classic meth synthesis pathways), is much more doable.
defense becomes much more difficult to maintain when you start throwing in out-of-distro tokens, breaking up the harmful uplift into benign chunks, and then piecing the innocuous-seeming facts back together, especially when you have jailbroken Opus helping you do it 😉
gg




English

If it's only visible to you, than it's usually pretty worthless, but not always. Some technical supports can impersonate accounts - which can lead to XSS execution (best case you can access credential and/or do CSRF).
But that would require social engineering and is less of a white hat activity
English


Aikido Security@AikidoSecurity
We detected a supply-chain compromise in onering 1.4.1, a Rust crate on crates.io with 18,000+ downloads. The latest version uses a malicious build.rs script to quietly exfiltrate git data and source code from your latest commit on every build, disguised as Sentry traffic. The GitHub repository is also compromised, so pulling directly from git is not a safe workaround.
ZXX

@GohansTips Vi que o download é 1,3 gb. Roda somente no PC e Android? Tenho o nds e switch
Português

💎Pokémon Eternal Emerald: Set in a climatically unstable Hoenn region threatened by an impending meteorite, a trainer from Johto must travel the region, challenge gyms, confront a new villainous team.
🎯Download Link Of Game -
gohantips.wixsite.com/pokemon/post/p…
English
Anatolij Vasilev retweetledi
Anatolij Vasilev retweetledi

@orenyomtov I not just making HTML slides? Who's gonna download and open PDF?
English

Anatolij Vasilev retweetledi

check this one
Run the npm login cli command
You will get a link like this /login/cli/uuid
Copy it and share it with the victim
The victim opens the link, and he only sees that the page is asking for otp, it doesn’t show him that his account gonna be logged in another session
When the user puts the 2FA code, their account will log in to the attacker's session
With good social engineering,it can be used to take over npm accounts
I reported this one 3 years ago and it still work xD
English

@thedealdirector After it hits 100% it will start social engineering
English

What happens when this hits 100%?
Asking for a(n insecure) fren.

Lisan al Gaib@scaling01
Claude Mythos and Fable 5 Benchmarks Anthropic is SOTA everywhere
English


















