[bitcoin-dev] Draft BIP for SNICKER

[David A. Harding]

On Sun, Oct 20, 2019 at 12:29:25AM +0000, SomberNight via bitcoin-dev wrote:
> waxwing, ThomasV, and I recently had a discussion about implementing
> SNICKER in Electrum; specifically the "Receiver" role. 

That'd be awesome!

> As the referenced section [0] explains, the "Receiver" can restore
> from seed, and assuming he knows he needs to do extra scanning steps
> (e.g. via a seed version that signals SNICKER support), he can find
> and regain access to his SNICKER outputs. However, to calculate `c` he
> needs access to his private keys, as it is the ECDH of one of the
> Receiver's pubkeys and one of the Proposer's pubkeys.
> 
> This means the proposed scheme is fundamentally incompatible with
> watch-only wallets.
> 
> [0] https://gist.github.com/AdamISZ/2c13fb5819bd469ca318156e2cf25d79#Storage_of_Keys

Your logic seems correct for the watching half of the wallet, but I
think it's ok to consider requiring interaction with the cold wallet.
Let's look at the recovery procedure from the SNICKER documentation
that you kindly cited:

    1. Derive all regular addresses normally (doable watch-only for
    wallets using public BIP32 derivation)

    2. Find all transactions spending an output for each of those
    addresses.  Determine whether the spend looks like a SNICKER
    coinjoin (e.g. "two equal-[value] outputs").  (doable watch-only)
    
    3. "For each of those transactions, check, for each of the two equal
    sized outputs, whether one destination address can be regenerated
    from by taking c found in the method described above" (not doable
    watch only; requires private keys)

I'd expect the set of candidate transactions produced in step #2 to be
pretty small and probably with no false positives for users not
participating in SNICKER coinjoins or doing lots of payment batching.
That means, if any SNICKER candidates were found by a watch-only wallet,
they could be compactly bundled up and the user could be encouraged to
copy them to the corresponding cold wallet using the same means used for
PSBTs (e.g. USB drive, QR codes, etc).  You wouldn't even need the whole
transactions, just the BIP32 index of the user's key, the pubkey of the
suspected proposer, and a checksum of the resultant address.

The cold wallet could then perform step #3 using its private keys and
return a file/QRcode/whatever to the hot wallet telling it any shared
secrets it found.

This process may need to be repeated several times if an output created
by one SNICKER round is spent in a subsequent SNICKER round.  This can be
addressed by simply refusing to participate in chains of SNICKER
transactions or by refusing to participant in chains of SNICKERs more
than n long (requring a maximum n rounds of recovery).  It could also be
addressed by the watching-only wallet looking ahead at the block chain a
bit in order to grab SNICKER-like child and grandchild transactions of
our SNICKER candidates and sending them also to the cold wallet for
attempted shared secret recovery.

The SNICKER recovery process is, of course, only required for wallet
recovery and not normal wallet use, so I don't think a small amount of
round-trip communication between the hot wallet and the cold wallet is
too much to ask---especially since anyone using SNICKER with a
watching-only wallet must be regularly interacting with their cold
wallet anyway to sign the coinjoins.

-Dave