Announcing Liquid.Coach

Today, we’re announcing Liquid Coach, a browser-based tool for Liquid developers and advanced users for doing operations like checking the balance of an address, creating and importing a PSET (Partial Signed Elements Transaction) or even sign it, all from their browser.

Features

• Load a blech32 confidential and bech32 unconfidential native segwit address
• Load an unconfidential wrapped segwit and legacy (confidential support coming soon!)
• Generate a mnemonic and derive an unconfidential native segwit address
• Select a custom Liquid explorer
• Provide an hex-encoded blinding key (if blech32 address) to unblind your utxos and watch the balance of your Liquid assets.
• L-BTC Faucet and Asset Mint on Regtest
• Create an unconfidential PSET with full control over inputs and outputs
• Add the Explicit Fee output
• Encode the PSET to base64
• Import base64 encoded unconfidential PSET and inspect and add inputs/outputs
• Sign with mnemonic or WIF

Network

The tool has been designed for developers and to mostly be used in regtest, although is possible to use it for Liquid mainnet too.

It uses a block explorer to connects and source blockchain data, it uses Blockstream.info for mainnet and an hosted instance of the regtest box Nigiri (tl;dr it adds on top of Blockstream Electrs a /faucet and /mint endpoint useful for regtest development along with automatic block mining)

You can use a local Nigiri as explorer, just select the Explorer button in the top-right of the page.

Limitations

Liquid.Coach is NOT meant to be used as a web-wallet for end-users. Use Blockstream Green instead.

There are few limitations, may or may not be fixed in the following iterations:

• You cannot load multiple addresses, just a single one.
• You can load a blech32 address and unblind the unspents, but you cannot create and sign a transaction with it. Cooperative blinding is kind of impossible with the current PSET spec.
• When you generate an address by default is unconfidential, so you can create and sign a transaction end to end.
• The generated mnemonic always uses the m/84'/0'/0'/0 derivation path

For any issue that may arise please open a ticket in the Github repository.