Suneal
@suneal_eth
Followers
343
Following
599
Media
14
Statuses
157
Security Researcher @zksecurityXYZ. Created @MetaMailInk. Prev WeChat.
Joined September 2018
Glad to have contributed to the security of the Solana ecosystem! Thanks to the @anza_xyz team for the swift response and timely mitigation. The bug involves a subtle Fiat-Shamir issue we call the Phantom Challenge. A full technical write-up will be published soon.
On June 10, 2025 a potential security vulnerability was reported to the @anza_xyz Github Security Advisory by @suneal_eth, a security researcher from @zksecurityXYZ. The issue has been mitigated on all Solana clusters and all funds are safe. More details below 👇
2
3
46
Bitcoin is an unstoppable agent that rewards those who help secure it.
1
0
0
Very interesting product. Just created my Circles profile! Looking to get trusted by someone who’s already in the network. @aboutcircles
https://t.co/F6g2w9t6Ml
Introducing Circles — an ambitious experiment on money. For the first time ever, you can create your own money without needing a bank or a government. 🧵
0
0
1
This is cool! LLM is changing how we review code.
We just shipped automated security reviews in Claude Code. Catch vulnerabilities before they ship with two new features: - /security-review slash command for ad-hoc security reviews - GitHub Actions integration for automatic reviews on every PR
0
0
2
This will be the best PlonK tutorial!
0
0
3
Trying to learn about the latest stuff in STARK-land? Come check out our new write up: Circle STARKs: Part III, Circle FFT https://t.co/nUTurQIcaM
0
9
32
We reported this bug to a number of projects (including semaphore) after finding it with our internal AI tooling (SnarkSentinel). The bug would have allowed proving that arbitrary leaves were part of a Merkle tree! Read more about how SnarkSentinel helps us find bugs 👇🏼
Public Announcement: 🚨 A bug was found in the BinaryMerkleRoot circuit in ZK-Kit. It allowed invalid Merkle tree leaves to generate valid ZK proofs. This has been fixed in v2.0.0 of BinaryMerkleRoot. @SemaphoreDevs v4, which relies on this circuit, is being updated
2
12
53
🧵 Over the last few months, we’ve published a range of blog posts exploring the technical challenges and ideas shaping zero-knowledge proofs, cryptography, and security. This thread rounds up the latest, from finding bugs using AI to post-quantum SNARKs 👇
2
7
34
This is another interesting ZK bug found recently. Read our blogpost to learn the query collision bug of Halo2.
We recently discovered a subtle but important soundness issue in Halo2, which we’ve named the query collision bug. It affects certain edge-case circuits and was present in widely used versions, including the main Zcash implementation and PSE’s fork. 👇🏼
0
3
14
Bug Hunt: Zero-Knowledge, Full-Paranoia, and the AI That Stares Back https://t.co/Bimbis4q6J
0
11
33
Just staked my selfie at @everstake_pool — 100+ chains, 0 downtime #EthCC8
0
1
2
Trust, But Measure: A Friendly Intro to TEEs with Intel TDX https://t.co/TVeeIV4blg For the 5th session of Proof is in the Pudding, we teamed up with Archetype to whiteboard an introduction to Trusted Execution Environments (TEEs).
0
18
76
We'll be in Cannes on Tuesday! Come meet the zkSecurity team and come talk about zero-knowledge proof, cheese, and wine over a ZK picnic! https://t.co/SdU2vnF1tP
1
6
26
@anza_xyz Here is our technical writeup about the vulnerability: https://t.co/83abzu16NM
Uncovering the Phantom Challenge Soundness Bug in Solana's ZK ElGamal Proof Program https://t.co/pO3X1hp0XA
0
0
1
@anza_xyz This bug started as a weekend curiosity and became a real-world disclosure. Thanks to the Anza and Solana teams for their fast response.
0
0
2
@anza_xyz We've published a full writeup explaining how sigma OR proofs work, what went wrong, and how to avoid similar bugs in zero knowledge systems: https://t.co/hefTV2m7yg
blog.zksecurity.xyz
In June 2025, we uncovered a serious soundness issue in Solana's ZK ElGamal Proof Program that could let attackers manipulate confidential token transfers undetected. We worked with the Anza team to...
1
0
4
@anza_xyz This bug highlights a tricky part of sigma OR proofs: Some "challenges" come from the prover, not the verifier. If these are not included in the transcript, the protocol can be forged. That’s why we call it the Phantom Challenge.
1
0
2
With this gap, a malicious prover could manipulate encrypted fee values to mint or burn tokens without revealing the actual transfer amount. We reported it to @anza_xyz through GitHub Security Advisory. They responded quickly and mitigated the issue across all Solana clusters.
1
0
2
The root cause was a subtle mistake in the Fiat-Shamir transformation. A prover-generated challenge wasn’t absorbed into the transcript. This broke the soundness of the sigma OR proof.
2
0
3
We found a critical soundness bug in Solana’s ZK ElGamal Proof Program. We call it the Phantom Challenge bug. It allowed attackers to forge sigma OR proofs and bypass fee checks in confidential transfers. Here’s how it works:
1
7
38