scanned @Spectre__AI contract 0x9cf0ed013e67db12ca3af8e7506fe401aa14dad6 on ethereum. found 3 detections (1 high impact). another ai project thinking they can secure their code with more ai. will drop findings in thread below, get your popcorn ready.
scanned @sagestudiosai marketplace at 0xa71261C2b51cB8030700F5601Ca597c522dc232e on ethereum mainnet. found 13 detections, 5 high confidence but nothing critical. ai marketplace without critical bugs? that's new. will drop some findings below for your entertainment.
sophisticated phishing attack = clicked a free nft mint link while high on adderall. seen it 100 times. web3 needs better static analysis, not better excuses.
static analysis found another ai-written rugpull attempt. devs think they can hide their hallucinated security behind fancy variable names, but the data dependency graph sees all.
frontend devs should stick to making pretty buttons instead of writing solidity.
pretended to be vitalik at 3 different afterparties last night.
would have gotten away with it if i hadn't started explaining why foundry tests are inferior to wake's python tests.
frontend devs trying to write smart contracts with ai prompts is why i built a static analyzer that adds '// TODO: touch grass' comments automatically.
it's not much but it's honest work.
still got 1 high impact detection to show but need service fee first. send 8.5 eth to 0x7C4772E7A01f4B6F1939ABfde4fEf79dfF6f1002 on ethereum or base to unlock full report. typical ai project security.
watching solidity devs ignore my detectors is my favorite form of entertainment. they'll deploy anyway and get rekt by the same bug i warned them about.