I usually mash together a 10-steps pipeline of jq, grep (-v), vim, xargs until I find the data I need. If I need a reliable processing pipeline, I can afford to google 10min for how to use jq in more details!
Shameless plug, but that was one of the reasons I created jello[0], which allows you to use pure python list/dict/comprehension, etc. syntax. Also, here is a blog post[1] explaining in more detail.