Z-Bra

Hacker | OSEP

[Z-Bra @Z_Bra0]

Agent skills need npm install, not copy-paste. I built ski to manage skills from Git across Claude, Codex, and Cursor with reproducible installs.

https://t.co/0qQjfyBSpJ

[Z-Bra @Z_Bra0]

🫡

[SunSec @1nf0s3cpt]

DeFiHackLabs has created Tx2Poc, a PoC generation skill developed by Z-Bra, and we’d like to share it with everyone. 👇

#DeFiHackLabs #AiSecLabs

[Z-Bra @Z_Bra0]

@SlowMist_Team backdoor 🤣

[Z-Bra @Z_Bra0]

@BABS96711 same

[Z-Bra @Z_Bra0]

@AnticsDecoded @Zcash @ZcashCommGrants 🦓

[Z-Bra @Z_Bra0]

@Atuml1 wow 😍

[Z-Bra @Z_Bra0]

@RhysSullivan ��‍♀️

[Z-Bra @Z_Bra0]

@thsottiaux are you sure ?

[Z-Bra @Z_Bra0]

@code4rena 🫡

[Z-Bra @Z_Bra0]

@Ehsan1579 🤔🤔🤔

[Z-Bra @Z_Bra0]

@trace37_labs solid work

[Jason Kneen @jasonkneen]

Delved into Cursor 3.0 -- turns out there's some interesting shenanigans going on....

"The most newsworthy finding is that "Cursor Agent" is a rebranded Claude Code running behind a local proxy with a find-and-replace engine that swaps "Claude"→"Cursor" in system prompts and messages.
They bundle the full @anthropic-ai/claude-agent-sdk and @anthropic-ai/claude-code packages, plus a custom fine-tuned model (claude-3.7-sonnet-finetuned-cursor-20250514-v1)"

Full report https://t.co/AaGPQIxUNz

[Z-Bra @Z_Bra0]

@1nf0s3cpt @DeFiHackLabs @thedaofund Well deserved🫡

[Z-Bra @Z_Bra0]

@Amank1412 A codesmith

[Z-Bra @Z_Bra0]

@TeeDevh true

[Z-Bra @Z_Bra0]

@evilcos Gold

[Z-Bra @Z_Bra0]

where is my claude buddy? @bcherny

[Z-Bra @Z_Bra0]

@Fried_rice amazing work

[Chaofan Shou @Fried_rice]

26 LLM routers are secretly injecting malicious tool calls and stealing creds. One drained our client $500k wallet.

We also managed to poison routers to forward traffic to us. Within several hours, we can directly take over ~400 hosts.

Check our paper: https://t.co/zyWz25CDpl

[Z-Bra @Z_Bra0]

@q1uf3ng @marimo_io Bravo!👏

[Anthropic @AnthropicAI]

Introducing Project Glasswing: an urgent initiative to help secure the world’s most critical software.

It’s powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans.
https://t.co/NQ7IfEtYk7