Quick Answer
I built this for people auditing their own contract before launch. Then I noticed bounty hunters were pasting Etherscan links into the wrong box — trying to triage other people's code, not their own. Here's the honest version of what this tool can and can't do for you.
Because it isn't true, and you'd figure that out in your first contest. The findings that actually pay on Code4rena and Sherlock are almost always something specific to how a protocol handles its own state or economics — a rebase token that breaks an assumption somewhere three functions away, a liquidation path that doesn't account for a fee. Pattern matching doesn't see that. Nothing that runs in under a minute does.
What pattern matching is genuinely good at: telling you, across forty files you've never opened, which ten actually have custom logic and which thirty are a fork of a library you've read a dozen times already. That's not a small thing when a Code4rena contest gives you a week and Sherlock's escalation process punishes rushed, thin reports. Time you don't spend re-deriving a function map by hand is time you spend on the two or three contracts that actually matter.
I'd rather tell you that straight than have you paste a real contract in, get a clean scan, submit nothing based on it, and decide the tool is useless. It's not useless. It's a triage pass, not a researcher.