It looks to me like it's very similar to a subset of sed, but with a more powerful regex library. But many programs in this language probably have near line-for-line translations into sed if they don't use newer regex features.
A lot of people (including me) have basically only learned the s command in sed and don't usually think about all of the more sophisticated sed features.
Yeah, it really is surprisingly similar to sed. "Surprisingly", because when I first designed the language I knew nothing about sed, and I still know very little (but I've seen people convert programs between both languages in a fairly straightforward manner).
A lot of people (including me) have basically only learned the s command in sed and don't usually think about all of the more sophisticated sed features.