A build may fail for two reasons: 1. Failure to build extension 2. Failure to run tests Splitting the status into build and test status makes a better indicator than a single general pass/fail. I still have no idea how to represent it visually though.