Offensive Security & Web3 Security Researcher
Joined September 2010
- Tweets1.7K
- Following192
- Followers192
- Likes4.5K
Guy retweeted
⚠️ TARGETED PREVENTIVE ALERT 🇪🇨: ALLEGED DATA EXFILTRATION FROM THE ECUADORIAN HEALTHCARE SYSTEM 6.6M patients
[STATUS: UNCONFIRMED; VISIBLE EVIDENCE EXISTS BUT NOT YET VERIFIED / SOURCE: TELEGRAM CHANNELS / DATE: SEPTEMBER 18, 2026]
Centralized cyber-intelligence monitoring has detected a post on Telegram "News" channels in which the threat actor "jarol488" claims to possess databases belonging to the Ecuadorian Healthcare System. To demonstrate the authenticity of the breach, the attacker has made a sample of 10,000 records publicly available.
It is strictly noted that this has not been officially confirmed; although visible evidence exists—
🛡️ TECHNICAL PREVENTIVE CONTAINMENT RECOMMENDATIONS (SOC / CSIRT / FRAUD)
Sample Forensic Analysis (Threat Hunting): Cyber-intelligence teams and EcuCERT must securely download the sample file (within a sandbox environment) from the provided link. Analyzing the data headers and structure will allow for the immediate identification of the specific database from which the leak originated (e.g., IESS AS400 system, MSP portal, etc.).
🖥️ CENTRALIZED THREAT MONITORING SYSTEM
Intelligence System: analyzer.vecert.io
Monitoring Console: monitor.vecert.io
#Cybersecurity #ThreatIntel #DataLeak #Ecuador #jarol488 #EcuadorianHealthcare #IESS #MSP #DataBreach #FinancialFraud #IdentityTheft #InfoSec #CyberAlert #VECERT #SOC #CSIRT #Unconfirmed #AllegedIncident #VisibleEvidence #SecurityAlert
Guy retweeted
We’re also introducing Gemini 3.8 Flash Cyber, our most capable cybersecurity model. It shows frontier-level performance in discovering vulnerabilities and patching them at scale, with Flash-level speed & pricing.
That includes achieving 86.2% on the important CyberGym industry benchmark, plus 47.2% on CWE-Bench for patching. We saw a 70%+ success rate in discovering vulnerabilities across 20 programming languages on our internal benchmark.
Guy retweeted
Claude code source code has been leaked via a map file in their npm registry!
Code: pub-aea8527898604c1bbb12468b…
Guy retweeted
Introducing Starknet Academy 🎓
The official learning hub built for developers who want to master Cairo and build on the most powerful ZK stack.
Guy retweeted
Mastering Ethereum (2nd Edition) — the go-to guide for Ethereum newcomers, builders, and even auditors.
If you want to truly understand how Ethereum works under the hood, this is where you start.
📖 Read it for FREE: github.com/ethereumbook/ethe…
Guy retweeted
Meet VulnLLM-R-7B: a specialized AI that reads code like a security expert. It's trained to spot vulnerabilities before they become breaches. This isn't just another chatbot, it's a digital security guard for your codebase. The community is buzzing because it makes security accessible.
Guy retweeted
New Engineering blog: We tasked Opus 4.6 using agent teams to build a C compiler. Then we (mostly) walked away. Two weeks later, it worked on the Linux kernel.
Here's what it taught us about the future of autonomous software development.
Read more: anthropic.com/engineering/bu…
Guy retweeted
ollama launch is a new command in Ollama 0.15 to run Claude Code, Codex, Droid and OpenCode with Ollama!
GLM 4.7 Flash is now optimized to use much less memory for longer context lengths (64k+).
Need additional hardware? Ollama's cloud offers GLM 4.7 with full precision and context length.
Guy retweeted
Hot Update!!!🥳
See you Today!
🤖 The $250K Bug: AI Security Research Meets Bug Bounties
Kicking off 2026 with two game-changers in Web3 security.
Gabriel Balko, a top Security Researcher (Known as @ControlZ_1337), recently earned a $250,000 bug bounty - the highest ever awarded for a vulnerability discovered using AI.
@MitchellAmador, CEO of @immunefi (The leading Web3 bug bounty platform), joins to share insights on how bug bounty programs work and what this AI breakthrough means for the industry.
What happened?
• AI agent found a critical vulnerability (preventing $500M+ in potential losses)
• 10 High/Critical bugs discovered in just 3 months
• Research that would take a year manually - done in a day
• 10,000x faster than traditional security research
Join us for a double session:
Part 1 – Mitchell Amador: Bug Bounties in Web3
✅ How bug bounty programs work in crypto
✅ What makes a vulnerability worth $250K+
✅ Getting started in Web3 security research
✅ AI's impact on the bug bounty landscape
✅ Industry insights from Immunefi
Part 2 – Gabriel Balko: The AI Discovery
✅ How the AI agent was built
✅ Translating human security intuition into automation
✅ Real results and stopped attacks
✅ Technical approach and tools
📅 Today!!! (Wednesday, January 14, 2026)
🕔 17:00 Israel Time (15:00 CET · 08:00 EST)
🌐 Online & Free
🎟️ Register: luma.com/m69hd5sn
📺 Live on YouTube: youtube.com/watch?v=j7IdANma…
This is where AI-powered security meets real-world bug bounties.
🔥 Powered by @MasterkeyVC
👾 Hosted by @Web3_Devs
#AI #CyberSecurity #BugBounty #Web3 #SecurityResearch #MachineLearning #Blockchain #InfoSec #AIResearch #Web3Security #CryptoSecurity #Immunefi #Automation
@AlonKivity @al23xyz , @ELuttenberg789, @Amit_Masterkey , @SHAP0W
Guy retweeted
First major hack of 2026, as @Truebitprotocol was drained for $26.2 million through an overflow in unverified bytecode.
The same attacker hit Sparkle weeks prior. Old code keeps bleeding - the archives have clearly become a shopping list.
rekt.news/truebit-rekt
Guy retweeted
A new report out from OX Security found that 900,000 users have had ChatGPT and DeepSeek conversations stolen by Chrome extensions.
That means full chat logs exfiltrated every 30 minutes.
Hackers want your AI history.
Why? Because you've probably pasted in code, strategy, legal questions, personal information, health info, or other material that could be used for blackmail or extortion.
A few takeaways:
1. Don't sign up for LLMs with your personal information if possible.
2. Don't input personal information into the LLM itself.
3. Be extremely careful with browser extensions. We already know that from wallet hacks.
Guy retweeted
Kali Linux 2025.4 Release (Desktop Environments, Wayland & Halloween Mode): Say hello to Kali Linux 2025.4! Expect updated tools, performance tweaks, and improved support - no fluff, just the essentials.
The summary of the changelog since the 2025.3… kali.org/blog/kali-linux-202…
Guy retweeted
ATTACK UPDATE: A massive supply-chain compromise has affected packages with over 2 billion weekly downloads, targeting *CRYPTO*
Here's how it works 👇
1) Injects itself into the browser
Hooks core functions like fetch, XMLHttpRequest, and wallet APIs (window.ethereum, Solana, etc.).
Ensures it can intercept both web traffic and wallet activity.
Guy retweeted
🍎🐛macos 15.6 is out, a few new CVEs and mentions...
NetAuth
Impact: An app may be able to break out of its sandbox
CVE-2025-43275
PackageKit
Impact: An app may be able to bypass certain Privacy preferences
CVE-2025-43232
CoreUtils
We would like to acknowledge...
Guy retweeted
Execution traces for Solidity tests have landed in Hardhat 3!
Together with inferred error messages and stack traces, Hardhat 3 offers the most complete debugging experience for Solidity tests.
Check out the Hardhat 3 alpha 👉 hardhat.org/hardhat3-alpha
Guy retweeted
El 2025 podría ser el peor año en robos de criptomonedas, con pérdidas que superarían los US$4.300 millones, según Chainalysis. En solo seis meses, hackers ya robaron US$2.170 millones, más que en todo 2024 y un 17% más que en el primer semestre de 2022.
trib.al/uUE5Oer
Guy retweeted
Meet the fully redesigned Solodit.
Cleaner UI. Smarter navigation. Access without an account.
More powerful for security researchers.
The same powerful data, now with a more streamlined experience.
👇
Guy retweeted
Los NFT u objetos digitales certificados han sido calificados de muchas formas: para sus críticos es una moda pasajera y no arte verdadero, para quienes los usan, es una manera de que la tecnología ayude al ambiente. prim.ec/bnug50VMFSa
