Wallet integration strategy

[PaulRBerg]

Problem

The MNTs and the streaming opcodes introduce a new blockchain paradigm, which won't be readily compatible with the existing web3 wallets.

While it will be possible to point to a Sablier JSON-RPC server and perform simple ETH payments, a vanilla EVM wallet won't be able to use any of the MNT or streaming features.

Note: this problem was pre-empted by @IaroslavMazur in this discussion. My discussion attempts to be a more comprehensive overview of the problem and of the potential solutions, all the while assuming the latest VM designs.

Potential Solutions

1. Collab

Ideally, we would manage to persuade an existing wallet developer to add support for SabVM. We work in parallel with them to bring SabVM to the market.

While I find this approach unrealistic, it may be worthwhile to talk to a B or a C-tier wallet, just in case.

2. Fork

This is the most plausible solution.

We will fork Rabby, or MetaMask, or Rainbow, and put the Sablier stamp on it.

Note

We have considered MetaMask Snaps at one point, but they don't work for us because token balances cannot be modified.

3. Custom Wallet from Scratch

Design a wallet from scratch.

I don't think @sablier-labs/frontend will smile upon this proposal, although maybe I'm wrong about this? I know you @razgraf said at one point that building an official Sablier wallet would be a great demo for SabVM.

Integration Points

• New Transaction Types. The wallet must be able to parse and send our new transaction types.
• Native Tokens.
  • The wallet must be able to read NT balances and make one-time NT payments.
  • Display all NTs being transferred during the tx signing flow.
  • We may want to create an official token list.
• Streams. The wallet must be able to show incoming and outgoing streams, as well as update the balance of the streamed NTs in real time.

RFC

CC @sablier-labs/sabvm

[IaroslavMazur]

@PaulRBerg, thank you for creating this discussion and offering a fresh perspective upon the problem!

---

I completely agree with the ordering of the potential solutions:

• ideally, we'd fully delegate the wallet-building to a dedicated team experienced in this (although, let's try our luck with the A players, too - maybe, we'll manage to hook one of them up on the idea! 😉);
• alternatively, we'd try building on top of an existing wallet solution ourselves; if this proves to be more problematic than implementing the thing from scratch, though,
• we'll have no other choice but to become experts in building web3 wallets from the ground up 😅

[PaulRBerg]

Great summary