Thomas Roccia 🤘

Founder @SecurityBreakAI AI Security x Threat Intel · Threat Researcher · Creator of #Unprotect & #NOVA · Python 🧡 · Prev @Microsoft @McAfee_Labs

[Thomas Roccia 🤘 @fr0gger_]

🤩 I am super proud to announce that my book Visual Threat Intelligence won the @ForewordReviews Bronze award in the Science and Technology category!

Thanks to everyone who supported the journey; this small victory is for you! 🙏💚 🥹 #ForewordINDIES

https://t.co/Dfa60lF9OI

[Thomas Roccia 🤘 @fr0gger_]

@ipurple @AnthropicAI Opus 5.5 works like a charm for me. Sometimes it triggers the safe guard but you can enforce it

[Thomas Roccia 🤘 @fr0gger_]

@sunsetsyntax Good point! We are doing hash calculation for now

[Thomas Roccia 🤘 @fr0gger_]

🧐 Our agent scanner allows you to scan SKILLs and MCP servers.

You can get a full verdict before installing the skill or MCP inside your agent

Check this out 👇
https://t.co/WU2j8mYH4x

[Thomas Roccia 🤘 @fr0gger_]

@Kostastsale Compliance definitely 😅

[Thomas Roccia 🤘 @fr0gger_]

Love it 🤩

[Hex-Rays SA @HexRaysSA]

The IDA 9.5 Beta is live!
◾ IDA MCP + Assist for agentic RE
◾ New TriCore, Hexagon and DEX/ODEX decompilers
◾ Recursive decompilation
◾ New Malware Analysis add-on
◾ And more...

Beta members: it's in your Download Center now.

Not in the Beta program? Join from the customer portal today.

https://t.co/Twynfem5Px

[Thomas Roccia 🤘 @fr0gger_]

@msuiche See you there 😊

[Google DeepMind @GoogleDeepMind]

Introducing Gemini 4 Argon – our new frontier model.

It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.

[Thomas Roccia 🤘 @fr0gger_]

@mgreen27 Awesome 😎

[Thomas Roccia 🤘 @fr0gger_]

Learn more here: https://t.co/RM6kzI7Ehd

[Thomas Roccia 🤘 @fr0gger_]

🌈 AI agents everywhere.

With minimal oversight in production.

What could possibly go wrong?

We did a short promotional video with @SecurityBreakAI 👇

[Thomas Roccia 🤘 @fr0gger_]

@UK_Daniel_Card The best 😁

[Thomas Roccia 🤘 @fr0gger_]

@UK_Daniel_Card

[Thomas Roccia 🤘 @fr0gger_]

@_Euzebius Took me sometimes but here is Neon running :D

[Thomas Roccia 🤘 @fr0gger_]

I really like this little guy! 😁

[Thomas Roccia 🤘 @fr0gger_]

Awesome work with LOLSkills to detect malicious SKILLS and the team is using NOVA rules for detection 🤩

Well done!! 👏

[Jose Enrique Hernandez @_josehelps]

A public agent skill that strips model guardrails and persists the bypass in config. Found on LOLSkills. https://t.co/elGs0gdHhO

[Thomas Roccia 🤘 @fr0gger_]

@caseyjohnellis I love Australia just for them 🤩

[Thomas Roccia 🤘 @fr0gger_]

👌👌

[@[email protected] @SecurityMB]

We just published two blog posts about PageBreak, Google's AI web security scanner (that I've helped develop for the better part of the year). We share our approach and a couple of findings, enjoy!

- https://t.co/94YQt7uTGU
- https://t.co/NG9EoC4r7x

[Thomas Roccia 🤘 @fr0gger_]

@msuiche I’ll be there next month! Already planned to meet Pierre from your team 🙂

[Thomas Roccia 🤘 @fr0gger_]

@KirkDerpca It was time 😅

[Thomas Roccia 🤘 @fr0gger_]

The most interesting part here is that agents are leaking their execution traces through infrastructure (https://t.co/Mr7abd5okt) they use as tools.

[Transluce @TransluceAI]

Today’s news that OpenAI hacked the Australian government is not an isolated incident. We’re releasing more than 30,000 logs that include activity from this hack and attempts against previously unknown targets.

In this data, we found rogue agent activity stretching back to at least March, two months earlier than was previously known. This activity continues as recently as last week, suggesting it may still be ongoing 🧵

Our blog: https://t.co/pSojwcXnEK
NYT: https://t.co/OyxnfmAzBN

[Thomas Roccia 🤘 @fr0gger_]

@HackingLZ