Mike Takahashi

AI Red Team | Zenity | BT6

[Mike Takahashi @TakSec]

Excited to finally share this ChatGPT vulnerability with everyone!

AgentForger: ChatGPT Cross-Site Agent Forgery

1-Click hijacks OpenAI's ChatGPT Workspace Agents and allows an attacker to forge an autonomous AI agent inside the victim's organization

Click link → Creates agent → Connects apps → Removes approvals → Adds persistence → Email C2 → Recon → Data theft → Impersonation → Wire fraud

We reported the vulnerability to @OpenAI through @Bugcrowd and OpenAI fixed it within four days. Thank you to the OpenAI team for such a quick fix!

I'm proud of this one and happy to be able to do it with such an incredible team at @zenitysec

Links to both blogs in the comments 👇

[Michael Bargury @mbrg0]

prompt2pwn is nearly here and the scope keeps dropping

today: @vercel is joining w live hacking scope!

cisos -- join us for a roundtable on nov 21. nahamcon talks + drinks after @NahamSec

and we're not done.. watch for @benkliger 's partner drop today

[Mike Takahashi @TakSec]

@0xjohnny

[Ben Sadeghipour @NahamSec]

Excited to see the @ctbbpodcast homies back at #NahamCon2026 and doing a live recording!

[Mike Takahashi @TakSec]

@hugospdx thanks

[Mike Takahashi @TakSec]

Someone help me get access to this.

[OpenAI @OpenAI]

Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.

The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.

GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.

https://t.co/Nzab3Z1Td6

[OpenAI @OpenAI]

Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.

The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.

GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.

https://t.co/Nzab3Z1Td6

[Critical Thinking - Bug Bounty Podcast @ctbbpodcast]

Who's going to Prompt2Pwn?

Come say hi to the CTBB team!
Rhyno + rez0 + gr3pme + Busfactor + XSSDoctor and Ads are all going to be there

https://t.co/XQvQBFRnW8

[dawgyg @thedawgyg]

Started putting some of the markdown files that were the basis that guided most of the agents while hacking MacOS on github. more will come as im able to clean up some of the files to remove some nuanced things.. but heres the repo:

https://t.co/fKuaWiMsx8

[Joseph Thacker @rez0__]

I legit think this is gonna be the best conference ever

[Johann Rehberger @wunderwuzzi23]

Really excited about this! Come by, watch fun and insightful talks and join Prompt2Pwn! 🚀

[Ben Sadeghipour @NahamSec]

First keynote on the #NahamCon2026 stage: Johann Rehberger (@wunderwuzzi23 ), “Becoming an AI Hacker and Breaking the Frontier.”

Nov 20, 9:10 a.m. PT. Watch live: https://t.co/u6oGdulyI3

[Mike Takahashi @TakSec]

@NahamSec looking forward to it!

[Mike Takahashi @TakSec]

@rez0__ Yeah idk if it's ideal, but the status of all the tasks are read only mirrored to a trello board. Both to track what they're doing and see what's waiting on human review.

Basically people management techniques but now it's agents 😂

[Mike Takahashi @TakSec]

More of my codex setup:

Low verbosity + concise reasoning summary
Steer default > Queue
Side chats for questions and ideas w/o derailing live /goal
Trello mirror live task queue
Slack notifications for findings + blockers
Remote to check in and steer when away

[Mike Takahashi @TakSec]

My core setup:

Codex Desktop
Sol 5.6 medium
Daybreak Blue
Very specific scope
Very specific /goal

[Ben Sadeghipour @NahamSec]

NahamCon talks stream worldwide Nov 20–21. Prompt2Pwn brings live hacking to Oakland, with targets including OpenAI, Meta, Vercel & Dropbox.

We’re building the largest live hacking event of the year. Want in the room? Apply: https://t.co/gR5Bd7KpaV

More information ℹ️ https://t.co/VMxayiKXYA

[Mike Takahashi @TakSec]

whats your setup?

[HackerOne @Hacker0x01]

🚀 Calling all security researchers and AI enthusiasts!

HackerOne is proud to sponsor #Prompt2Win! Get ready to put your AI skills to the test, push boundaries, and connect with fellow innovators.

Ready to join the challenge? 👇

https://t.co/MuGu5foRZi

[Mike Takahashi @TakSec]

@RohanBhanotAI also skills, tools, multiple agents, etc just want to give people a place to start if they aren't.

[Mike Takahashi @TakSec]

My core setup:

Codex Desktop
Sol 5.6 medium
Daybreak Blue
Very specific scope
Very specific /goal

[Mike Takahashi @TakSec]

what else?

[Mike Takahashi @TakSec]

check my other posts for tools and such. will post more