It lets you write anything you'd write with normal RegExps, but with a more readable syntax, and it lets you compose sub-expressions into larger patterns.

It also provides helpers like:

- `atomic()` that prevents the engine from backtracking into what you passed as argument. No more ReDOS.

- `bound()` works like `\b` but for arbitrary character sets. E.g. `bound(/[_\p{L}\p{Number}]/u)` is a Unicode-aware `\b`

- Set operations on character classes: `charSet.intersection(/\p{Lowercase}/u, /\p{Script=Greek}/u)` will only match lower case greek letters

I hope you'll find it useful!