Jeff Security (@jeffsecurity) profile picture

Independent Smart Contract Researcher & Researcher at @ShieldifySec My mission is to find vulnerabilities in smart contracts for a safer Web3 Space!

Joined May 2021
Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)Photo by Jeff Security (@jeffsecurity)
Made $60k last month from audit 😎 What about you?
23
142
11,052
🫣🫣
October 29th 2026 will mark the bottom for crypto. Bookmark this.
2
420
🚨 Binance suspected hack: $5.5B+ drained, the largest crypto theft ever (Bybit was $1.5B) What we know: • 20:24 UTC: Binance 14 hot wallet sends 167K ETH, 1.6B USDT and 60+ tokens to a fresh address in one tx • 3 min earlier: a Tornado Cash funded address installed an EIP-7702 delegation on the hot wallet • Total outflows: ~$3.5B on Ethereum, ~$1.44B in BTC, ~$630M on Tron • Tether froze 1.57B USDT, Circle froze 823M USDC (~$2.4B total) • Binance: investigating, says user funds are SAFU. CZ: “it does look like a hack” The real concern: ETH, BTC and TRX can’t be frozen. Watch for laundering via DEXs and mixers. The 7702 detail is the one for devs to study. A delegation persists and can authorize later transfers without another signature. Not confirmed by Binance yet. Follow official channels only and beware of scams. Track the hacker wallets: arkm.com/explorer/entity/21a…
2
8
1
23
2,116
1/4 EIP-7702 can turn the wallet itself into the sweeper. A Binance-labeled wallet moved 167,464 ETH + 63 tokens, including 1.615B USDT, just 3m24s after a signed delegation was installed. This is bad. etherscan.io/tx/0x53bd38b816…
48
137
65
679
252,448
Jeff Security retweeted
Security is more important than ever for this space to scale. Please integrate AI into your workflow, but never fully rely on it.
Our teams are aware of reports of a potential security issue and are actively investigating. User funds are completely safe. Our security teams are reviewing all relevant information with urgency, and we will provide updates through official Binance channels as soon as we have verified information. Users should remain cautious of scams and should not trust unofficial links, messages, or requests for account information.
2
6
293
Groth16 verifier bug: gamma and delta in the verification key were set to the same G2 point 🙃 They must be independent. When they match, the pairing equation collapses and valid-looking proofs can be forged with no deposit and no witness That is how Foom Cash lost $2.26M in March. The cause: one snarkjs Phase 2 setup step was never run Math was fine. Setup was not rekt.news/the-unfinished-pro…
1
4
13
680
The next bull run will test every protocol with real volume and real attackers. Fuzz it, formally verify the invariants that hold your funds and get it audited. Launch like you expect to be attacked, because you will be 🫡
1
3
11
580
Jeff Security retweeted
Daml bug pattern worth knowing: a Canton app checked allocation.admin == registry but never required the registry to sign. Anyone can implement Token.Allocation. @JJS_OnChain breaks down his first Daml critical 👇 nitter.cf/JJS_OnChain/status/209…
2
2
15
1,237
Daml bug pattern worth knowing: a Canton app checked allocation.admin == registry but never required the registry to sign. Anyone can implement Token.Allocation. @JJS_OnChain breaks down his first Daml critical 👇 nitter.cf/JJS_OnChain/status/209…
2
2
15
1,237
New shieldify.org is live 😱 Request an audit in a few clicks, and browse 170+ security reviews filtered by chain, language and status. shieldify.org/audits 🫡🫡
5
6
17
809
Jeff Security retweeted
rini is live in beta. First vault managers can now deploy modular vaults on HyperEVM, with access to Hyperliquid perps, spot, HIP-3 and HIP-4. It's a beta, open to whitelisted managers only.
15
17
14
57
4,439
Solid resource 😱 This is one of the most structured open-source prompt packs I’ve seen for bug bounty hunting! @mdp_sec broke down their entire workflow into 10+ reusable phases: → Recon & surface mapping → Authenticated access setup → Endpoint & vuln-class sweeping → Browser-based walkthrough → Critical-impact hunting → Triage, dedup & validation → Independent verification gates It’s Web2 focused, but the methodology is gold for anyone building AI-assisted security workflows. The evidence-first, phase-by-phase structure is exactly how you avoid hallucinated findings and actually produce reliable results: github.com/mdpsec/bug-bounty…
4
20
83
3,176
This deep dive on multi-agent software engineering pipelines shows how to partner with AI to build production-grade, highly tested Web3 systems. Watch the full talk: youtu.be/UcZdy0JRrCQ
1
9
671
Jeff Security retweeted
Updated $DRKW on DexScreener and CA added to website. We're also audited by Shieldify (@ShieldifySec). Audit 👇 github.com/shieldify-securit…
5
9
1
18
2,266
Jeff Security retweeted
Another audit report for @pear_protocol is live ❇️
Another audit report for @pear_protocol is live ❇️ We reviewed their PEAR → PEAR migration to HyperEVM: a one-way LayerZero bridge, rate-limited minting, staking migration with preserved timestamps, and the HyperCore spot listing. Read the full report 👇 github.com/shieldify-securit…
1
3
15
452
Another audit report for @pear_protocol is live ❇️ We reviewed their PEAR → PEAR migration to HyperEVM: a one-way LayerZero bridge, rate-limited minting, staking migration with preserved timestamps, and the HyperCore spot listing. Read the full report 👇 github.com/shieldify-securit…
2
7
2
22
1,323
Hash-perfect decompilation of a Solana program is now a thing. @osec_io went from a 220KB .so blob to byte-identical Rust source for GoonFi V2, using agents scored on how closely each function matched the original bytecode. The bug they found once the code was readable? The AMM priced assets off its own vault balance. Donate tokens, skew the price, drain the other side. $6.1M at risk. If your program is live, assume your source is public. osec.io/blog/goonfi-decompil…
1
4
1
11
780
TryHackMe’s AI Security path: 25 hands-on labs on prompt injection, jailbreaks, AI supply chain attacks and RAG data poisoning. ~19h, intermediate. Worth it if you audit anything with an LLM in the loop. tryhackme.com/path/outline/a…
3
1
17
690
“Tier 1” security firm vs. @ShieldifySec: audit results on the same codebase. Choose wisely 🫡
3
13
871