On July 25, we hacked OpenAI. Two bugs let us take over ChatGPT/Codex accounts of OpenAI employees (+some unaffiliated users) and reach connected services: Outlook, Slack, GitHub, etc. We proved it with a PR in OpenAI’s internal codebase . It took us <72h. 🧵

We know you're all hyped with DEF CON stuff but here's one more thing to get excited about: @J0R1AN just released a new research over at the Critical Research Lab! And this time it was inspired by a discussion we had on our discord, go check it out! lab.ctbb.show/research/solvi…

A vulnerability in Azure? How about 3 of those? Uncovering 2 SSRFs + a file upload path traversal I discovered in the Azure API Management service. Attackers could send requests from the service’s proxies, access internal Azure assets, deny service and upload malicious files🧵

Breaking Claude Code Opus 5 Auto Mode 🔥 1/ Here is a somewhat hilarious attack chain that hijacks Claude Code Opus 5 for a full system compromise via a website Hint: Security invariants are not optional 🧵

🚨 CosmosEscape: We found a single key that unlocked every database in Azure CosmosDB One key >> Platform-wide impact. Microsoft fully remediated the issue. ✅ Read the full research and see how Wiz uncovered CosmosEscape: wiz.io/blog/cosmosescape-tak…

The whitepaper is live! Read "Can AI Do Novel Security Research? Meet the HTTP Terminator" here -> portswigger.net/research/htt…

JavaScript Sandboxes: A Small and Casual Antipattern Review jamvie.net/posts/2026/07/jav…

Seems that wp2shell PoCs are now floating around the internet, so we've published our blog post including our research methodology for finding the bug as well as a deep dive into the chain itself - slcyber.io/research-center/e…

MAD Bugs: My Cousin Vinyl (CVE-2026-50052) So the story went like this: Squid was bleeding from a 29-year-old heap overread in her default config. Naturally, she did the only sensible thing: she called her cousin, Vinyl. It turns out Vinyl also had a family problem of his own. blog.calif.io/p/mad-bugs-my-…

Pleased to publish a browser-related research paper (w/@inzo____) titled: One trigram at a time: XSLeak via Universal CSS Injection and DoS in Opera (GX) The title speaks for itself. Enjoy the read! zhero-web-sec.github.io/rese…

On June 10th, we announced a critical auth bypass in phpBB, now CVE-2026-48611. Here's the technical followup with exploit scenarios and how to detect it. All it takes is one unauthenticated request to log in as any user, admin included, on a default phpBB install, no password required. Aikido Attack caught this on a routine run. We reported it June 2nd, and phpBB shipped a fix four days later in version 3.3.17. If you haven't upgraded yet, do it now.

We've published a new blog post by RyotaK @ryotkak He discovered 8 methods to bypass safety mechanisms in Claude Code, leading to arbitrary command execution. We recommend updating to v1.0.93 or later to fix this vulnerability (CVE-2025-66032). flatt.tech/research/posts/pw…

The written version of my BSides Riga and @bsidesvilnius talks is up: exploiting git integrations in cloud services, with four bugs I found in GCP (Looker, Dataform), including the one that won me MVH. nopnop.pro/2026/06/17/exploi…

👨🏻‍💻 Did you know that it’s possible to perform RCE in Internet Explorer via clickjacking? Igor Sak-Sakovsky's (@Psych0tr1a) new article will explain how! swarm.ptsecurity.com/the-cli…

New short article on a real-world exploitation case rather than pure research, demonstrating how a specific mistake in Next.js can lead to a systematic zero-click SXSS on its latest versions (w/@inzo____): Re:CACHE - Excessive reflection, type confusion, and 0-click SXSS on Next.js zhero-web-sec.github.io/rese…

We've published a new blog post by RyotaK @ryotkak ! He exploited a directory deletion race condition in Google Cloud's Looker, leading to full RCE and K8s privilege escalation. Read the technical details here: flatt.tech/research/posts/re…