python-shamir-mnemonic VS slips

You can recover with regular Linux/MacOS/Windows computer using this python script https://github.com/trezor/python-shamir-mnemonic

I suggest you do some further reading here https://github.com/trezor/python-shamir-mnemonic/issues/15

You can use Electrum or any third party wallet to enter the seed phrase. Otherwise with Shamir use https://github.com/trezor/python-shamir-mnemonic/

For even more security, shard the mnemonics with Shamir's secret sharing.

I do hope multisig gets easier in the comming years, but just the same, Shamir could also be standardized (the trezor team has already done the work and has the proposal and implementation) in the same way.

I mention that in the issue. The problem is this rationale seems unjustified because a 12-word BIP-0039 mneumonic encodes 16 bytes (128 bits) + 4 checksum bits of information before the passphrase comes in https://github.com/trezor/python-shamir-mnemonic/issues/40

However, the trezor reference implementation only allows to be given an existing secret in the form of hexadecimal seed key and not the actual mnemonic, as provided.

Trezor support Shamir backups (at least the model T does), but the cold wallet itself should be stored in the safety deposit box with the bank too (because if you know the device password and have the wallet with you then you are vulnerable to a wrench attack). You can generate one for an existing seed phrase with this command line tool also: https://github.com/trezor/python-shamir-mnemonic/

The official SLIP-0039 standard itself confirms it is not possible to convert this mnemonic type to BIP-0039. Down in Section 9 "Compatibility with BIP-039":

For anybody new or returning to SSS, check out SLIP-0039: https://github.com/satoshilabs/slips/blob/master/slip-0039.m...

One of the big downsides of SSS is that it’s very raw and you have to do a lot of legwork to make it actually useable. It’s rightfully criticized for this and the argument follows the don’t roll your own crypto vein.

SLIP39 solves this by formalizing a protocol for handling SSS splits built atop standards for crypto key serialization (BIP-39). SlIP shards are unique on each generation so parties with the same underlying SSS shard can’t compare mnemonics, they’re mnemonically serialized, and have a checksum and group index metadata which makes a more sane UX possible when combining.

Well every wallet chose to solve this problem independently. Trezor proposed a new standard called SLIP10 to do BIP44 type operations coins that did not use secp256k1. Problem is, there are very few utilities I've found that will do SLIP10 derivations.

Checkout the SLIPs repo (https://github.com/satoshilabs/slips.git) and modify testvectors.py. We are going to replace the curvenames and last four show_testvectors lines with the following:

They use an open source algorithm which is documented here. Anyone can verify it and the recovery outside of a Trezor hw-wallet is possible if required.

It will be using SLIP-39, like Trezor and Electrum, or a Ledger rewrite of it. All the language about shards is straight from the SLIP39 spec.

This was all done with the SLIP-14 seed using the passphrase coinjoin if you want to follow along.

With the introduction of the new Coinjoin feature in the latest release of firmware and software, I had the need to dump some of my taproot derivations. Although blockbook can do this fine using descriptors in place of xpubs for taproot accounts, it fails on Coinjoin accounts. This is likely because SLIP-25 as 6 deep derivations while BIP-86 uses a standard derivation depth of 5.

The recovery of Coinjoin accounts is described here. Accessing them outside of Trezor Suite will 100% destroy all privacy obtained since Suite is the only keeper of the anonymity set for each UTXO. Using your CJ coins outside of Suite may also erode the privacy of previous transactions using your Suite Private coins as well.