When I was first starting out, the first alg I came up with was R U2 R' U' R U' R'. I used it to do a 3-cycle of edges on the last layer.
I found a set of G-perms myself when I was learning PLL - R' U L' U2 R U' L F B U2 F' B' - and the mirrors and inverses of course. I also came up with M2 U M2 U' M E2 M E2 for Z-perm, and r U r' U' r' F r2 U' r' U' r U r' F' for A-perm.
When I was learning BLD, I found the above poster's alg for orienting 4 edges.
When first learning square-1, I came up with /3,0/-3,0/3,0/-3,0/ (just a sexy move basically). Then I move-cancelled that with /3,-3/-3,3/ to get a pure J-perm /3,0/-3,0/3,0/0,-3/-3,3/, which I used for speedsolving for a long time.
I'm sure there are plenty more I can't think of right now. This isn't counting algs I've generated, numerous commutators, fingertricks for existing algs, or entire methods, of course.