If I create new solutions, libraries then is often my task to search for existing solutions to use them or use them as library to create faster and more solid the goal. The current status is more feeling, a mind set on my side to decide is this repository useful and still active in maintenance. So I look at the project is it a 'one man show' or a lot of committers, an organization. This entirely depends on my priorities and field. The more experimental or niche work my do. Different languages may also exhibit different technical traits here which makes this more or less of a concern. In the end I will have to decide for myself, but a list of helping points is more transparent and supporting my decision.
So I want to create clues and indications (not really rules) to get an easier decision to use or not use this library in my project.
Maybe you had also thought about or you have the same problem.
How you decide about the dependencies in your project? What's your rules for or again usage of an existing solution?
Thanks to all your time!
Below is a list (in no particular order) of some traits I look for.
Recentness, tests and source code quality are probably the major deciding factors for me, assuming an API meets what is required or can be extended to meet additional needs.