Seeking Feedback: Would a Fedimint-Based Escrow Module Be Useful?

[Harshdev098]

Hey I'm working on a potential Fedimint escrow module and trying to validate whether it could solve a real problem before investing more development effort.

The current design is as a buyer to lock ecash into an escrow contract specifying a buyer, seller, arbiter(s). In the normal case, the buyer releases funds to the seller. On a dispute or after a timeout, the arbiter signatures will be valid to sign the outcome to Release the funds to the seller(if seller wins) or refund to the buyer(if buyer wins) the decisions made will be offchain. The federation enforces contract rules and signature verification.

I've put together a short demo showing the workflow: https://youtu.be/-My9o9cfBRk
I'd love your thoughts on the workflow itself rather than the implementation details.

A few things I'm particularly interested in:
Does this workflow fit your use case?
Where would it break for your users?
Is there any information would buyers, sellers, and arbiters need that I'm missing?
Would a single-arbiter model be sufficient initially, or would multi-arbiter support be important?
What would need to exist before you'd consider integrating something like this?

Also i think it would be interoperable with nostr native applications we can either use Pontmore(https://github.com/pontmore/protocol/) or any nips that is feasible just the arbiter discovery event needs to be there for that kind!

cc: @ManfredKarrer @ripcurlx