Elliot Belt

Had the opportunity to present at @INCYBER_Europe on a topic I've been working on: using agentic MCP pipelines to speed up detection engineering and purple team cycles. If you're interested in the technical side, I wrote a detailed article about it: itrust.fr/purple-team-qu…

Last week happened the THCon 2026 CTF. A huge congrats to all our participants, especially the winners of this year on-site CTF : - 🥉 1984 (@UT3PaulSabatier) - 🥈 FlagaDOS (@ITrust_CyberSec) - 🏅 Old Tokenz (various companies)

I was just awarded the AI ninja badge on HTB Academy! academy.hackthebox.com/achievement/ba… #hackthebox #htbacademy #cybersecurity

Spent the last few months going through research papers on LLM-assisted security work. Compiled everything into one article: structured prompting, false positive reduction, multi-stage pipelines, and the real numbers that make your LLM optimal: felixbillieres.github.io/posts/promptin…

Just completed the "Dojo #51 - DeadBolt" challenge on @YesWeHack! Level up with me! 🌟 dojo-yeswehack.com/challenge/play… #YesWeHack #ChallengeAccepted

I just solved Love on Hack The Box! labs.hackthebox.com/achievement/ma… #HackTheBox #HTB #CyberSecurity #EthicalHacking #InfoSec #PenTesting

@wepfen @yeswehack Who is US twin🫩✌🏼🥀

@Elliot_belt @yeswehack gg twin

Just got a reward for a vulnerability submitted on @yeswehack -- Information Disclosure (CWE-200).

Leave some for us 😭😭😭 (Or answer the Claude Advocate form I filled so I can join 🙄😏🫰(thx again @_mudpak 💌))

🔓 Just beat the "Dojo #49 - Secret manager" challenge on @YesWeHack! Think you can match my skills? 🌟 dojo-yeswehack.com/challenge/play… #YesWeHack #ChallengeAccepted

Recursive Language Models (RLMs) let agents manage 10M+ tokens by delegating tasks recursively. This Google Cloud Community Article explains why ADK was the perfect choice for re-implementing the original RLM codebase in a more enterprise-ready format →goo.gle/4kjT12E

The Anthropic Fellows program provides funding and mentorship for a small cohort of AI safety researchers. Here are four exciting papers that our Fellows have recently released.

After quite some time researching MCP protocol (@AnthropicAI) for offensive security projects, I worked on building a little learning path covering core MCP features, offensive & defensive tools for anyone wanting to explore this space. github.com/felixbillieres… @alexalbert__

You may have heard about OpenClaw – the most viral open source project of the year. It's an agent and messaging gateway that lets you automate digital tasks through platforms like Discord. In this course, you'll learn how it works along with key security practices like using Docker-based sandboxing to protect your host system while your agent executes complex workflows. freecodecamp.org/news/openclaw-…

"Dojo #48 - RubitMQ"? Pwned! It was a blast on @YesWeHack! Think you can take it on? 🌟 dojo-yeswehack.com/challenge/play… #YesWeHack #ChallengeAccepted

Why I don't trust MCP output (and neither should you) Not a cool research, just some small notes and thoughts blog.slonser.info/posts/smugglle…

I just pwned Redelegate on Hack The Box! labs.hackthebox.com/achievement/ma… #HackTheBox #HTB #CyberSecurity #EthicalHacking #InfoSec #PenTesting

I just pwned Phantom on Hack The Box! labs.hackthebox.com/achievement/ma… #HackTheBox #HTB #CyberSecurity #EthicalHacking #InfoSec #PenTesting

🔓 Just beat the "Dojo #47 - APICrash" challenge on @YesWeHack! Think you can match my skills? 🌟 dojo-yeswehack.com/challenge/play… #YesWeHack #ChallengeAccepted