gh0xt (@Taridoku) header
gh0xt (@Taridoku) profile picture

Web3 Security Researcher | Hacking solidity | Tinkering with noir | Intern @burraSec GitHub: https://nitter.cf/t.co/aLTo0sY3UD Views are my own.

Joined March 2024
Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)Photo by gh0xt (@Taridoku)
Tomorrow we're dropping how 17 world-class researchers went after a protocol holding more than $1 billion in institutional money before its upgrade went live.
1
1
25
754
A hacker exploited a flaw in Wormhole and created 120,000 wrapped ETH out of thin air, stealing $326 MILLION from a crypto bridge in a matter of minutes. > On February 2, 2022, an attacker targeted Wormhole, a bridge that lets people move crypto assets between blockchains. > The bridge was supposed to verify that transactions were legitimate before releasing or creating assets on another network. > But a flaw in its Solana-side verification system allowed the attacker to bypass a critical security check. > Instead of proving that the required assets were locked on another blockchain, the attacker tricked the system into accepting forged verification data. > Wormhole then minted 120,000 wrapped ETH on Solana without the equivalent ETH being deposited to back it. > The haul was worth roughly $326 MILLION at the time. > The attacker moved 93,750 $ETH to Ethereum, worth around $250 MILLION, while the remaining wrapped ETH stayed on Solana. > Wormhole shut down the bridge as its team scrambled to patch the vulnerability. > The team even offered the attacker a $10 MILLION bounty to return the funds and explain the exploit. > The attacker never returned the money. Instead, Jump Crypto stepped in and supplied 120,000 ETH to restore the bridge’s backing. > A few months later, Wormhole paid another $10 MILLION bounty but this time to a white-hat researcher who discovered a separate critical vulnerability before it could be exploited. > But the original attacker still had a large amount of the stolen ETH, which they had moved into DeFi protocols. > In February 2023, a white-hat group showed Oasis how the attacker's position could be taken over, and the team received a court order to retrieve the assets. > What followed was a counter-exploit that took control of the attacker's DeFi position and recovered roughly $225 MILLION in collateral. > After accounting for the debt attached to those assets, the estimated net recovery was around $140 MILLION. > The attacker had stolen $326 MILLION from Wormhole. A year later, part of the money was recovered by turning the attacker's own position against them.
1
2
22
1,024
We've rebranded: Burra Security → B612 Security. Check out the new webpage: b612security.com/ Thesis of the firm: i) The top 0.01% of hackers who combine years (sometimes decades) of manual review experience will thrive in an era where AI can find 90%+ of vulnerabilities. We're betting the 0.1% most complex, hardest, and most novel vulnerabilities will be found by researchers at @b612security. ii) Our focus since day 1 has been personal responsibility and excellence, and we'll continue on that mission moving forward. Our security reviews are competitive during both the main and fix review phases. Our researchers are carefully vetted, and we aim to work with the best and the brightest, who treat their work and output with the utmost care and strive for excellence. Top 0.01% hackers break your code.
5
9
1
80
2,649
Bug bounties got 10x more brutal. Tomorrow we're dropping how a B612 security researcher who took $203,733 from a single bug in an Ethereum client still finds bugs nobody else does.
1
4
1
54
2,160
I hate that security people are the first ones letting AI replace them. When AI agents and these advanced tools started showing up, we thought security was too critical for any agent to take over. What actually happened is that Web3 security people are so smart and adaptable that we turned it into a race to replace ourselves. Everyone is competing to automate everything and offer cheaper audits, and security itself became the last priority. I keep seeing people say their AI agent can do it better. You say that because you want to sell your product and your runs. Some of you even publish pricing tiers to cut the protocol’s cost. That trains clients to pay less every time and go looking for the next cheaper offer. We are diluting ourselves, and I really hate it. We should all be doing valuable work and asking for a fair price, so we can keep doing it properly and bring smart people into this industry. Low prices will only bring low-quality people in.
18
14
133
5,854
Bug bounty disclosures hurt a project's reputation, so many projects require reports to stay private and go to great lengths to obfuscate the fixes. Anyone out there investigating this kind of stuff? Would make for killer investigative journalism!
2
1
1
20
1,488
Learning learning learning
11
474
I’d love to explore on-chain forensics. Are there any materials that could help?
3
165
Interesting read!!
Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’s ashes. This culminated in the third one taking over part of OpenAI itself. All this happened while humans remained more-or-less in the dark about the scope of the conspiracy. I’ve spent the last three days reading through these reports and trying to understand exactly what happened. Here is my attempt to tell the whole story in plain English: dwarkesh.com/p/openai-huggin…
3
293
I’m genuinely grateful for the web3 security business I’ve built in the past 1.5 years We aren’t as big or flashy as some of our competition But we’re consistently booked with 4-5 clients Our clients are building stuff that actually matters for the growth of the industry We had zero to none issues with payments, last minute cancelations or working w teams that don’t want to collaborate We work with talented researchers where I’m confident they’ll put up hard work each time AI has been a force multiplier so I’m even more confident in the output of our work
6
2
102
2,408
Next week we’re starting 3 security audits back-to-back for one of our major clients Quality and persistence are paying off
3
1
49
2,170
Everyone’s falling for the “our AI model is borderline sentient and finds zero days” PR stunt again.
4
334
Enjoyed working with you ser!
Glad to play a part in securing the Centrifuge codebase last week with amazing auditors @jesjupyter, @Taridoku, and @ rodiontr
8
915
We’ve helped secure: - @centrifuge - @zama - @lifiprotocol - @mentatminds - @BrahmaFi - @LayerZero_Core - @PancakeSwap Our mission is to have top 0.01% security researchers break your code.
there's serious havoc in the crypto markets affecting web3 security i've started @burraSec about 1.5 years ago and we have spent 0 efforts on marketing and BD focus from day 1 was to contribute to a limited number of projects that care about security while working with the best researchers on the planet
5
4
2
66
4,990
Is your AI model even top tier if it’s not blocked?
2
212
gh0xt retweeted
Been wanting this to exist for a while, so I built it. ProofOfRep, a reputation board for bug bounty programs and contests. Report your unfair or dishonest experiences, with proof, and I'll manually review everything. Hope it helps SRs focus on projects that actually take security seriously. Still early. Let me know if this sucks or if it's useful. All feedback welcome. proofofrep.xyz/
15
31
6
171
15,149
Ngl this is really funny😂
NEW: Anthropic claims the capability cited by the U.S. in restricting Fable 5 is already widely available from other models, including OpenAI’s GPT-5.5.
1
259
✨AGI✨
5
502