
Solidity || Sui Move Security Researcher || Formal Verification || π§βπ» SR at @QuillAudits_AI || My opinions are personal
Joined October 2021
- Tweets1.1K
- Following448
- Followers403
- Likes4.9K
Had the chance to test QuillShield early, and it's a real shift.
Instead of waiting until the code is "done" to find out what's wrong, every PR gets reviewed with full protocol context from day one.
Security that moves at the speed of development. If you're building, grab a beta spot.
8 years of auditing smart contracts. Thousands of findings.
And the same pattern every time: the code was done, the audit was next, the launch was waiting.
We built something that reviews the code while it's still being written.
First look at QuillShield.
The most powerful AI security reviewer for onchain code we've ever built. It reviews every pull request with the full context of your protocol, starting from a complete audit of it.
Opening a small beta. DM for access.
Another audit wrapped at @QuillAudits_AI . 14 findings: 2 High, 6 Medium, 2 Low, 4 Informational.
Alhamdulillah, just put together this simple setup for auditing.
Nothing fancy, but it's where the bugs will be found. Time to lock in. π
Iβve seen so many beautiful setups, MacBooks, and clean workspaces on my timelineβ¦
This is mine.
Just a simple laptop. No desk at all. No gear.
But every day, this is where I learn, struggle, and keep going.
Some days are hard.
Some days I feel behind.
But I donβt stop.
One day, Iβll have that dream setup.
But more importantly β Iβll earn itβΊοΈ.
For now, this is enough. π»
Another audit wrapped at @QuillAudits_AI . 10 findings: 1 Critical, 5 Medium, 4 Low.
If you price WBTC with the BTC/USD feed, you have encoded an assumption you never test: that a wrapper and its underlying are the same asset.
The day that stops being true, your oracle is still correct. Your staleness checks still pass. And your liquidators lose money on every position they touch.
Ask your team what happens at 15% divergence. If the answer is "that will not happen," that is the finding.
π€ Made with AI
400 followers. Thank you to everyone who has read, replied, and pushed back on something I got wrong.
Proud to have been one part of this.
The number I keep coming back to is 12 projects delivered on time. Anyone can find bugs given infinite runway. Finding them on schedule, every month, is the actual discipline.
August was one of those months where you just sit back and feel genuinely proud of the team.
Our audit team @QuillAudits_AI absolutely crushed it: β
12 projects audited - all delivered on time
π 155 findings identified
Huge kudos to @kalp_eth, @Pro___King1,@phoenix244001, @cryptanu, and @turvec_dev for the exceptional work and relentless commitment to securing Web3.
Hats off to you all. On to the next one....
Pro King π retweeted
August was one of those months where you just sit back and feel genuinely proud of the team.
Our audit team @QuillAudits_AI absolutely crushed it: β
12 projects audited - all delivered on time
π 155 findings identified
Huge kudos to @kalp_eth, @Pro___King1,@phoenix244001, @cryptanu, and @turvec_dev for the exceptional work and relentless commitment to securing Web3.
Hats off to you all. On to the next one....
This time, the codebase was relatively small, six medium-severity, one low-severity, and four informational findings.
π€ Made with AI
Pro King π retweeted
If you're an auditor feeling stuck, watch @zigtur talk.
Feedback Loop: Every missed bug is only wasted if you don't feed it back into how you hunt the next one. That loop is what separates auditors who keep improving from those who stall.
youtube.com/watch?v=_5Fya4Luβ¦
Pro King π retweeted
One week audit, wrapped. Team engagement at @QuillAudits_AI .
4 High, 4 Medium, 2 Informational.
One week audit, wrapped. Team engagement at @QuillAudits_AI .
4 High, 4 Medium, 2 Informational.
Two week engagement wrapped, this time as part of a team at @QuillAudits_AI . 2 High, 2 Medium, 3 Low, 1 Informational.
Results from last week's private audit of mine at @QuillAudits_AI 2 High, 3 Medium, 1 Low, 2 Informational.
Enjoyed this one a lot. Deep, uninterrupted review time on a well-scoped codebase is the best part of this job.
π€ Made with AI
Wrote a formal spec for a client's contracts during a private audit last week. One property failed. The counterexample turned out to be a high severity finding.
The prover does not get tired, does not skim, and does not assume. More auditors should have it in their toolkit.
Can someone please help me get TAC for @OpenAI? I'd really appreciate it. Thanks!
Multisig users: run a quick scan with Multisig Inspector before your next signing session.
A safe contract signed blindly is still a loss waiting to happen.
Introducing Multisig Inspector π₯·
$1.7B+ lost across multisig incidents in the last 18 months. Almost none of it was a smart contract bug. It was someone signing something they didn't fully verify.
We built the tool that closes that gap. π§΅




















