Motivated by the question Would a userscript to determine whether a review is an audit be illegal, I wrote a userscript for this purpose. The algorithm is as follows:
If the post score shown in review is 0, the script checks its real score using the API. If the score is greater than 1, the post is a known-good audit. If the score is less than -1, or the post does not exist, it is a known-bad audit.
The difference of 1 vote is not considered conclusive, since there is a remote possibility of someone up/down voting the post in just the right time. The known-good audits actually have score 5 or higher, and known bad (it seems) have score -2 or less.
The audits are indicated by the color of the review bar.
Known bad audit:
Known good audit:
The logic of suggested edit audits is different; the script is not tested in that queue, but probably fails to detect audits accurately.
