It seems like a few of you have already noticed, but I've implemented account verification, which checks your auth to make sure you are who you say you are. It requires the download of
an add-on. I'll make it so unverified accounts aren't able to do much after I get back from my final this afternoon. I also started to lay the foundation for the peer review system. If you're logged in and verified, you'll be able to see that there's a new button on the error reports while viewing add-ons. It's currently not functional, but will be relatively soon.
I'd personally prefer an add-on approval system like what RTB had over a forgetton of false positives
There's an add-on with literally no code (a print) that is being detected as using eval
For now, I'm just going to try this out. I'll select a small handful of people who can instantly 'greenlight' and add-on and moderate to get rid of spam and duplicates, but I'd like to leave it open to all to help contribute to reviewing add-ons.
Currently, the mark-up system and scanning are entirely separate. The scanning is very rudimentary and scans for the very mention of any of the "problem words", regardless of where it is placed in the syntax. I'm going to merge the two systems as the mark-up seems rather proficient in separating out the syntax, so it'll be a much more intelligent system and unlikely to call out false positives. Currently, it's finding the term "call" within the word "automatically", which is part of a comment. It's very clear to me that this is extremely broken.