There's still a code review to be done by another person for quality assurance. However, the finishing touches to h5p.com are currently binding Joubel's resources. The review is on the list of issues that will be tackled next though, so it won't take "forever". You can track the progress in Joubel's ticket system.
Regarding the score: Did you set the mastering percentage to 75 % in the behavioral settings? Then the maximum score will also be set to 75 % of what's possible, too. It would be weird if you wouldn't have a full score bar and a filled star on mastering -- H5P computes the score bar based on the achieved score and the maximum score.
Also, this behavior was set by design: Essay is not Fill in the Blanks or a word guessing game, but rather a less-than-ideal solution for roughly estimating how well a text is. That's why setting a mastering score lower than 100 % means that it is not necessary to have included all the words. Using this approach students won't have to include all of them to master a task. That's a different notion than having "only" found 75 % of the score(d words) but "generously" being granted to have mastered the task. You wouldn't grade an essay by comparing it against a list of keywords either.
Furthermore, because Essay is not really capable of assessing an essay, it has the option to deactivate the scoring completely and just provide some feedback based on the keywords given or not given.
I have elaborated on that issue some more in a blog post (in German).
Well, after having typed all that, I hope that setting the mastering grade to 75 % was actually the reason, and there hopefully is not a bug