Djinns have a lot of design pitfalls

I am not a build-maker. I will not be referencing specific stats. This is mostly expressing my feelings trying to build and level a Disciple of Varashta as soon as possible, leaning mostly into Djinn commands and supplementing the rest of the build with skeletal minions since they generally share stats. I focus heavily on minion dps and cooldown. Only some went into survival since Djinns can't die, but skeletals can.

Ruzhan (Fire Djinn)
The main djinn I tried to build around. The built-in command skill feels okay damage-wise. The main problem I have is trying to build around it requires sacrificing your other command skills for this. Particularly, the "Short Fuse II" makes this skill very nice to use with its almost half second detonation time, but using that support is a detriment to all other of Ruzhan's support because the downside affects all skills, not just the detonation skills. This points to a deeper problem with how Djinns are built where all command skills share the same supports. Any specialty supports is an opportunity cost or worse a detriment to other commands.

Trap: Feels decent for mobbing, but fairly RNG with rune placement. Looks really cool, but clutters the screen.
Fury: Playing on controller, this skill misses like 70% of the time I use it. On stationary bosses like the giant skeleton boss, Ruzhan usually directs this skill based on his last orientation, not based on where I am targetting. While it looks cool and seems like it would be good damage, I simply cannot use it because of how often it will miss.
Reckoning: Nice for additional damage, though the long cooldown makes it harder to instinctively use since it is probably on cooldown when I try to use it again. It also does not feel very responsive due to it having to teleport before performing the attack.

Navira (Water Djinn)
Having a support-oriented djinn is really nice to have as it is more universal to build with. The built-in command skill feels slow to execute for when it matters. In theory, it would be nice to use it as cover for my skeletal minions, but whenever I use it, it was usually out-of-place, not actually covering anything that matters. Using it never felt like it mattered.

Fracturing: Did not use. Since I was focusing on Ruzhan, I was not building for chilled ground so that I could use this skill.
Well: It feels like an alternate to Convalescence. Shorter cooldown, but arguably weaker effect (uninterrutable recharge is stronger) and longer realistic cast time. Due to it being a command skill, the time between issuing the command receiving the buffs is noticeably long where it could affect my survival.
Oasis: Likely due to my inexperience, but the benefits of this is not something I would ever take. For an ascendency point, I do not value refilling flask charges. Recoup as life does not help my ES build (that side of the tree normally builds for ES anyways). Recoup as mana does nothing for me since I do not have mana problems (flasks fix that problem). Finally, this requires chilled ground to use which is an additional hurdle. For me, there is no situation I would take this command skill to use, even if I had a different build.

Kelari (Physical Djinn)
This command skills available for this Djinn was very good throughout the campaign and good in maps. The critical weakness makes it so even the one point to take the djinn is great since its built-in command skill is good at applying that debuff. The stacked usage makes it comfortable to use in every encounter, letting it quickly recharge to full stack when moving between packs. It appears to apply several debuffs due to how the projectiles work which makes is usable for applying a decent stack during a fight.

Malediction: Mainly restricted to mobbing where corpses are plentiful. Tried to theorycraft sacrificing skeletons to fuel the beetles, but gave up since it my dps is being capped by the corpse consumption rate. It looked like a lot of work for a test and I am not into build-crafting to this extreme. Maybe someone else can make it work. The one thing this skill DOES do well is it being a separate skill to socket supports into, making it easier to build the djinn.
Judgment: Did not use since I was not building for any kind of armor break/reduction to properly utilize physical damage. Looks cool. Hoping the damage is worth the full charge-up time.
Deception: Amazing skill in conjunction with Kelari's built-in command. This combo was my main leveling skill in campaign. I alternate using these two for mobbing so both have the opportunity to recharge. I can also spam this for high burst damage against bosses. Feels easy to use and effective in most situations.

Overall, the main problem with Djinns is being forced to build around one command or generalize the supports so it benefits all commands. Specialty supports are hard to justify if you have multiple commands. Aside from Kelari, most command skills do not feel great to use.

Side note: The passive tree feels limited for scaling for minions/djinns dps. Trying to plan out a build, I notice there were not many minion/command damage clusters.