jferrant

Currently the signer and miner do some minimal checks to prevent a single malicious signer spamming the block. This should be improved. In theory, a signer can only vote once...

Currently malicious signers are only reported via a block rejection message in the case of SignErrors. The signers do not report malicious activity that happens during DKG rounds. I think...

There's an inner message protocol at work when calculating the "vote" bytes of a signature. This really should be captured into its own struct which hides the details of how...

Nakamoto Signer[3.0] - Coordinator selection revamp

3

Currently coordinator selection is done with a zeroed consensus hash just to enable tests to pass . There needs to be a revamp of the selection logic. The chain tip...

Currently stacks-signers assume that a DKG round will be successful after they have broadcast their vote. However, a signer need to check if a vote failed for a specific round...

Nakamoto Signer[3.0] - Add timeout logic for when to trigger a new DKG vote

2

The stacks signer needs logic to trigger another DKG round if a certain amount of time passes and we still have no confirmed DKG vote (maybe a certain number of...

Currently if a signer crashes, it is not enough to simply retrieve the DKG key from the contract Use the wsts save/load functionality to persist state needed to continue signing/dkg...

Signers currently spam the node on start up to see if the reward set is calculated yet. Instead, have it only query the node once on startup. Subsequent reward set...

Currently signers will sign blocks in the order that they arrive. However, if a miner proposes a block while they are mid signing another, it means they have given up...

Currently the stacks signer spams the endpoint at various points in the code base. A lot of this information should not change often and should not be queried on every...