A project's README file provides human readable information about the project
such as a basic description, instructions for getting started, and how to
contribute. But this file is not typically optimized for parsing out these
individual pieces of metadata. How then should projects expose this data in a
machine readable way to power things like project directories or automated
discovery of related services like continuous integration?
This project is designed to document and discuss existing solutions to this problem, identify gaps, and collaborate on any new work that may be useful in this area.
A successful result of this effort would be a consistent, agreed upon method of expressing project metadata in a vendor-neutral way than could be utilized by a variety of consumers.
See prior art.
Discussion also happening on the TODO mailing list.