Thanks for posting the draft. Can you please elaborate a bit what your ideal result would be?

As you mention Renaissance masters: Do you want to adopt their style and create something similar? In that case, it would be worthwhile to study a book specifically devoted to Renaissance counterpoint. I only know de la Motte (very inspiring and highly recommended), Daniel (very rigid and much less motivating, but the section on the modes is worthwhile reading), Jeppesen (quite outdated), and Fux (if you want to brush up your Latin (sic!), although the idea of "species counterpoint" is at odds with the practice of Renaissance music). I guess there are other, more modern books in English (I have heard positive comments about the book by Peter Schubert). Although it is, in principle, possible to extract all the conventions by observing a lot of original music, you can spare a lot of time by relying on the authors of counterpoint books who have already done this. Even if your goal is to deliberately deviate from this style, it is a good idea to learn the historic style so that you know where you deviate and maybe can even use such deviations as expressive means.

Or do you want to create music in a different style that is more or less reminiscent of the style, but clearly in a different idiom? Something like Rheinberger's Mass for double choir or Bruckner's "Os justi"?

Thanks for the hint to Athanasius Kirchner's specification of the composition rules. The video is somewhat sensationalist, but it includes a link to a sober description of a software implementation by Andrew Cashner.

I am not sure, though, whether Kirchner's method from 1650 actually is an algorithm, because it leaves decisions to the user. On p. 16 of Cashner's paper describing his implementation, he explains that he added a "badness" measure that is optimized for making these decisions. It thus seems to me that Kirchner described a grammer rather than an algorithm. It would require some research, though, to investigate whether it actually can be reformulated with a formal grammer according to Chomsky, of which type of the Chomsky hierarchy this grammer is, and maybe even devise an algorithm that decides whether a particular piece of music can be generated with this grammer.

Note that this is not the only grammer describing (a subset of) Renaissance music. An earlier example is William Bathe's table from 1596 for constructiong canons over a cantus firmus. Moreover, all the rules given in counterpoint textbooks can also be seen as a grammer for a musical language, that is, a subset of all possible pieces of music. The most comprehensive modern try to achieve this for Renaissance music is Thomas Daniel's counterpoint book, which is unfortunately put down by interspersals of arrogant qualitative judgements about music examples, which are quite embarassing for the author (for instance, he apparently is not aware that a soggetto used by Zarlino is the "Ave Maria" and calls the soggetto "stiff and falling flat towards the end" [p. 19]).

These grammers will help you creating some music that would not have raised 16th or 17th century eyebrows, but this is not an end in itself. Firstly, you must have something to say. For sacred music, this implies that you have some religious aim in mind. There is a wide variety of such aims (e.g., the music of Bach and Telemann had a very different purpose from that of Palestrina and Victoria), but you should first ponder this for yourself.