VulnerabilitiesrUs

Welp, I found out that, this is a HVCI Compatible BYOVD. That's even scarier. This means it will work even on Windows 11. Proof below @0XDbgMan @weezerOSINT @horsicq @thezdi Expect a well throughout POC. CVE Incoming! x.com/TempAccountNul…

@jahirsheikh8 Oooh sassy! I like it. It does have a brain!

At this point, Claude Fable 5 has run out of patience 💀

@kernelstub This seems very fishy to me. However, it seems to me that you’re attempting something like this to chain vulnerabilities to do something. Guessing based on your posts.

Hi! I am looking for RCE 0days for apache, nginx, ssh, ftp, react, and way more :). There is no budget. You can contact me at my Tox. D9630B228389292FA71AB36958C1EEF9751834A401DAC8D053A591E1511AE968778B9E44C1E3

I'd like to propose @OpenAIDevs / @thsottiaux a modification to the Usage Limit functionality for "ChatGPT Codex". Current Problem: When the 5-hour usage limit is reached, users are blocked even when they still have plenty of weekly usage remaining. The current workaround is to use a full reset, but that feels wasteful because it resets both limits when only the 5-hour allowance is exhausted. I’d like to propose a proportional option for banked resets, so users don’t have to spend a full reset just to recover from the shorter 5-hour limit. For example, each banked reset could be used as either: - 3–4 five-hour resets, or - 1 full reset covering both the weekly and current 5-hour limits. Once that’s in place, an optional opt-in Auto-Use setting could automatically apply a five-hour reset when the 5-hour limit is reached.

Let me just be real here a moment. Every program does not need an Electron interface! I'm so glad you chose DirectX here, it looks so much better and on top of this, you have full control of your window, no chrome, no electron, no asar issues, no npm, just pure rendering through GPU->dxgi!

I’ve been fighting the “finish something” problem hard lately. Today I'm making GitSmarter public as my first open-source repo. Native Windows Git client. C++. Direct2D. ~2.5 MB. No Electron. Implements Git itself instead of wrapping it. Still early. Proud of it anyway.

Yes. IF you don't want to risk getting your account banned and consistently getting it unbanned after re-appealing each time. Then absolutely. But please keep in mind, it does not guard you against other things. It MUST be strictly cybersecurity/Legitimate Disclosure purposes only. Otherwise, it will remove you, after `3` flags, and completely ban your account as you didn't abide by the Cyber Trust Rules and Regulations.

@_watzon @saltyAom has it helped?

???????

@saltyAom Yep! Businesses/enterprises must use this!

Has anyone applied to this? I ran a deep security scan on my library and got redirected to this page, but it seems to be Enterprise? Just want to know if any OSS devs got approved through this before I apply

We’re releasing new Qwen3.6 quants that run 2.5× faster on your GPU. Qwen3.6-27B NVFP4 runs on 24GB VRAM. 35B-A3B can hit 17,561 tok/s (B200). We also improved accuracy, tool calling, agent use, and looping. Guide: #nvfp4" target="_blank" rel="nofollow noopener">unsloth.ai/docs/models/qw… Qwen3.6 NVFP4: huggingface.co/collections/un…

@eliteobsolete What? If I was going to eat my Mac, and cause irreparable damage to it how does the second line matter ??? You literally destroy it 😂

I’ve noticed that Many people are complaining about limits being burned on both codex & Claude. Here’s the issues I have found with testing. - Having too many enabled MCP tools. These will be big requests, alongside your prompt. IF you have a lot of them and when you say ‘continue where —‘ This adds up over time per model. Try disabling or removing some that you likely never use. - Not actually using a prompt to your question before sending the request. For goals - Not giving enough instructions so you stop eventually. Either your question, or goal is not giving it more instructions with what you are trying to accomplish. Or, it’s trying to think for you based on memory.md and what it knows otherwise, It will guess and check itself when you aren’t descriptive even stop during the way; if say something fails, cant understand, or you lose connection. Always prompt what you’re doing for, ask it in your 5-hour quota to build the skill with all these resets it will be able to prompt, request the right tools, not confusingly guess and check itself, and burn through less of your quota. Love you guys, happy research.

I miss those days when programming discussions were fun

Yes, that's Counter Strike on a PSP! Yes, that's DevTools inspecting game UI! Yes, it's PocketJS! Clean room implemented FPS engine, fully open JavaScript mod API, ~12MB RAM footprint, 60fps, fully open source. ⚡️

Big news from @OpenAI today: starting September 1, all individual members of Trusted Access for Cyber (TAC) must enable Advanced Account Security using a hardware-backed passkey to retain access to frontier cyber models. AI safety is nothing without identity assurance. Learn more about our partnership with OpenAI, and how TAC members can now secure their workflows with a custom 2-pack set of YubiKeys at preferred pricing. 👉 yubico.com/blog/openai-ma…

Nebula Security is now backed by Y Combinator. We’re celebrating by bringing you the world’s first Android 17 root demo — “IonStack”, a url click can let attacker fully control your phone. This is not only an Android root demo. We’re bringing you a full chain browser-to-kernel exploit with two 0-day vulnerabilities affecting Firefox before v151.0.2 and all Linux distros in 15 years. "IonStack" demonstrates how bad actors can control your phone by sending a malicious URL, but good news, Nebula Security found it before attackers do. Both 0-day were found by our code scanning agent, VEGA, overshadowing any vulnerabilities found by Mythos or any scanner you name it. VEGA has demonstrated its extraordinary capability in finding critical bugs in the world’s most complicated software: operating systems and browsers. It can spot the same vulnerabilities in your codebase too. VEGA support full scan and incremental scan that can integrated into your CI/CD flow. We launched VEGA within YC companies and received overwhelmingly positive feedback. Now it is open to all enterprise customers in private beta. Book a demo with us: nebusec.ai/book-a-demo

@Yubico @OpenAI This key is such a lifesaver! I have one for everything I use! Never trust Mobile/Email 2FA.

Chat experience in the new ChatGPT app is a total mess - No Projects (idk where they are now) - New chat opens in a side popup - If you want it fullscreen, it opens a new clumsy window - Text zoom doesn’t work - Global shortcut opens a task, not a new chat - Chat history is no longer in the sidebar It feels like they just duct-taped Chat onto Codex

Razer, MSI, any big brand that ships a flash drive or hardware it is baked into the chip. Once you plug-in that drive it runs the autorun normally as it's attestation signed. Same with Disc's/Dvd's in the past. Now, with hardware, like GPU's or HID's this is baked into the chip itself. It auto launches as soon as windows recognizes it with Windows Update.

Some LG, Alienware, and Dell monitor owners have discovered that simply plugging their monitor into a Windows PC can automatically install adware without asking for permission first. Users are mad because the app immediately starts showing McAfee pop-up ads after it’s installed. Apparently, this happens because Windows automatically downloads companion apps for supported devices, while Windows handles the installation, monitor manufacturers decide what the app does after it’s installed.

Lets gooooooooo!

Google Earth Pro 26 years old what a shame 👋🔪💀

The new ChatGPT App should have a dedicated Chat mode alongside Work and Codex. Not every interaction is about productivity or coding. Sometimes people just want to think, explore ideas, learn, vent, create, or have a casual conversation. A simple option like: Chat For casual conversation would make the app feel more complete and much closer to how many people actually use ChatGPT every day. ChatGPT is not just a work tool or a developer assistant. It is also a conversational space. @ChatGPTapp @TheRealAdamG @thsottiaux @JustinBleuel @jxnlco @Dimillian @AriX @ajambrosino

here's hoping! It would genuinely help with research disclosures!