Thanks for the update: the situation seems to be clear then and the planned actions seem sutiable to address it: if the update is out with the changes I wil lbe happy to test it out with the exact same scenario again

There is a 4x game that I keep in very high regard and where the AI plays diplomacy very satisfyingly and they have also gone into similar direction: being friendly or hostile to all the individual factions influence your dealings with them

This is how they have done to create an intuitive mechanic and avoid loopholes:

1. The perception of the other factions does not matter directly (it would lead to an endless loop that is hard to track) only concrete diplomatic actions and contracts have effect: non-agression pact, trading, declaring war and who has declared war etc...

2. When dealing with a faction all the modifiers that have been generated from the above are listed on the screen when dealing with them e.g:

+ 5: We are trading (short time and/or low intensity trading)

+ 20: We have common enemy (supported in war over a long time or several times)

- 5: You are trading with our enemy

- 10: You have declared war on our ally

3. Random events have influence in these dealings to prevent the player to fully have control over the situation (random diplomatic incidents can occur at random intervals)

4. Whenever someone declares war on another faction they ask all the other factions they are fiendly to or allied if they support them. Support yields positive, rejection yields negative perception

5. If ally is not supported in war, the alliance might be broken right away (but the better the perception the more likely that the alliance will remain in place)

6. Both positive and negative modifiers scale over time to a maximum value meaning factions rank up huge bonus of malus due to ongoing war or war support and thus alliances and even complete alliance blocks are formed organically: the more some factions fight against common enemies the more likely it will be that they will ally and or support each other in more wars and get even better opinion of each other and vice versa :)

7. All the modifiers degrade over time thus the player cannot build up guaranteed support right in the beginning: an unexpected war where the player ends up on the other side can distance them form former allies if not put an end to in time as historically good relationship degrades naturally, negative perception is being built up though constantly during current conflict. Sme is true the other way around and even former enemies can be won over if their interest allignes later as negative perception fades away and new positive bonuses bring them together

This particual game also made every single diplomatic action cost influence: here we have free actions but over 1-2 actiosn we need to pay influence which I guess would balance things out

Would need ot keep an eye on the free actiosn not being spammed

Made some recommendations on diplomacy itself:

https://www.games2gether.com/amplitude-studios/humankind/forums/169-game-design-and-ideas/threads/46764-diplomacy-improvements?page=1#post-345108

As for vassal: their independence war woudl be the perfect opportunity to allow other factions to support them

Otherwise they will not have the power to truly gain independence and would only be able to pose a minor annoyance to a powerful faction

If external powers support them they can allign and gaining independece be draw into the powerblock that has supported them