Already 2 years ago, i wrote about how easy it can be to write Apache Sling Pipes to Extract, Transform and Load your content. In the team i am working on we have been using them pretty extensively, with mostly positive feedbacks.
However some people were not yet plainly satisified because of two reasons:
1. it was still a bit difficult to apprehend what you can use, how and when,
2. when you want to validate and archive the pipes you've run against your system, what you were storing in the git repository was the FileVault representation of its "compiled" persistence. This makes reviews, change, history checking really tedious.
3. Also my estimate colleague Jörg Hoh asked for more / better documentation.
Ultimately a project sold to help translate a few words between sling developers into few words that can be executed on a sling instance was ending up in a complex bunch of stodgy XML files.
Finally pipes find their original meaning
with release of Apache Sling Pipes 4.1.2, you can now POST simple pipe commands as text, or as a file containing one or more pipe commands. And with adobe.com's open scripts package, not only you have all dependencies, and new security settings set for you, but there is also a terminal client here to help you execute that servlet, called pipe.
then it comes easy and intuitive to search and modify content (using jq here to put JSON in form)