User:Moonythedwarf/extra-unreliable

This is a Spam/PR wording/phrases highlighter script that attempts to identify phrasing that does not constitute an encyclopedic article. It is not perfect, and it is not to be used mindlessly. It does not understand context. It cannot read, and may highlight perfectly acceptable text. (albeit rarely) It is not perfect, see limitations below for details. The script cannot tell the difference between a perfectly acceptable userpage, and blatant spam, that is up to the user.

This script uses regex rules to attempt to match on phrasing and words commonly seen within marketing material, spam, and low quality articles.

Due to the strategy used by this script, it cannot read, cannot understand context, and is otherwise imperfect. It works by attempting to detect specific patterns and highlighting them. This pattern recognition ability is limited to what Javascript's regex implementation can provide. As of now, it is also easily confused by pages dense with formatting, and may fail to match on text it otherwise would detect.
Alongside this, the way matching is currently implemented means custom descriptions and formatting does not work on a per rule basis, and this may take some time to be fixed (as it's a complicated problem)