[Group-Init][Turn-Marker] When are these two getting married? And a suggestion for [Group-Init] ...

I just think the two should totally be one API, I can't see a reason one would want one and not the other! The suggestion, and more than probably a very extensive one, is involving the ever so dandy 5e Surprise Rules.  Would it be possible to incorporate surprise when determining initiative?  Say something like, those who are marked with the Ninja Mask (which in my game is not a single stealth check, but a Hide Action, stealth is the contest rolled within view of eachother) are attempting to surprise the enemy.  The flow would then go like this: 1. Determine Line of Sight of each in contest, query obscured conditions in chat window to GM.  Make sure blindsense, darkvision, tremorsense, and normal vision ranges are observed. 2. Autodetermine who is trying to stealth (has ninja mask icon) and roll a stealth check from their character's sheet, get that variable 3. Autodetermine/ roll perception check(s) of those in line of sight, determine who wins contest, mark/ store any surprised for later in script 4. Create the main stack of initiative as rolled. 5. Create and use a new stack of initiative that places a "! " before the names in this stack of initiative order for those who are surprised, move them to inaction this round  (the "! " being the marker to skip their turns in combat).  This also lets those in combat know that there is advantage on their rolls against said combatant.  In turn marker, instead of "Round 1" being listed, the name could be changed to "Surprise Round". 6. Once the turn is over and surprise is no longer a issue, end this stack of initiative, call the previously made stack and continue as normal.

Some games dont use initiative systems that work well with Group-Init, or using that script might be more than a particular GM wants for their campaign. Turn Marker works fine for people who don't use Group-Init. Plus, your example is a very specific use case, and group-init is specifically set up to be usable in lots of different game systems, while Turn Marker is completely system agnostic. It might not be possible to handle such a specific sequence without losing their flexibility for other games. Steps 1-3 are extremely complex, and would really be worthy of a script in their own right, and would only work for one game system.

G G said: Some games dont use initiative systems that work well with Group-Init, or using that script might be more than a particular GM wants for their campaign. Turn Marker works fine for people who don't use Group-Init. Plus, your example is a very specific use case, and group-init is specifically set up to be usable in lots of different game systems, while Turn Marker is completely system agnostic. It might not be possible to handle such a specific sequence without losing their flexibility for other games. Steps 1-3 are extremely complex, and would really be worthy of a script in their own right, and would only work for one game system. Point conceded ... Yes, perhaps a 3rd API then is called for.

Yeah, a third script would be ideal. There’s a concept in software engineering called the Single Responsibility Principle which states that a particular system should have only one reason to change, meaning one thing it is responsible for. You can apply that at many levels, but for GroupInitiative, it’s “Roll initiative for a selection of tokens” and for TurnMarker it’s “Highlight the token whose turn it is.” Some other system needs to provide “check surprised state for all combatants.”  As G G points out, that would likely be very Game System specific, or at least apply to a subset. There might be some facilities that the other two could expose beyond what they do to make it easier to interface the three, but it definitely belongs outside them. 

There are some very excellent scripts in Roll20 that I do not use, simply because they do not follow this principle. Particularly when script that purportedly is built to do X, also does Y, Z, in a way that interferes with script that does Y (and does it better).

gotcha 3 on all points.  But what do you think about doing up a surprise roll for d&d 5e that ques up group-init, or performs the same functions as group-init afterward, that is 5e specific?  Or as an option to the existing group-init that can be included in the methods of determining initiative, that is off (by default) unless specifically wanted?

or perhaps, in all three cases, making a stand alone api that merges these three apis, or each seperately - creating a suite, if you will