Published Nov 14, 2022
By Josh Goodbody, COO Qredo

Decentralized Custody Manifesto I: Self-Custodial Exchange Integrations

Trust, but verify.

A timeless mantra, but also one in dire need of a refresh. 

Where did this saying come from? If you believe Wikipedia, it’s from a Russian proverb, but it was actually made famous by Ronald Reagan as he grappled with non-proliferation of nuclear weapons.

For years we’ve been compelled to trust centralized service providers. We see enigmatic CEOs, leading from the front, signaling all the correct virtues, and yet, time and time again, this trust has led to disaster.

Trust. Can’t Verify.

And yet despite the demand of trust, could a user of a centralized service actually verify much?

Well, you could identify the founder, the leadership team, and perhaps dig into a few company filings and check LinkedIn. In short: superficial verification.

Most entities in crypto are private companies, with financials largely hidden and ownership interests tangled up in a web of offshore jurisdictions with little to no corporate governance requirements. 

So what are they doing with your assets? How do they store your crypto? Do they commingle your funds with theirs? Are they lending or trading with clients assets? 

Users have no means of checking this.

Lessons Learned. Finally?

The dramatic collapse of FTX is the proverbial cherry on top of the pile of centralized entities that have failed their users.

Celsius, BlockFi, Voyager, 3AC, Axie Infinity, Terra —  they all have one thing in common: an explicit requirement that users trust them. 

Centralized exchanges seem to have woken up to the reality that cryptographically provable mechanisms for verifying trustworthiness (typically called  “Proof of Reserves”), aren’t a nice to have, but a necessity to survive. This is progress. 

Nevertheless, I would argue that centralised exchanges are living on borrowed time.


At some point it is going to become so blindingly obvious that exchanges should not be trusted to secure and govern client assets.

It doesn’t happen in traditional markets, and it shouldn’t happen in crypto.

From here, two models will emerge:

Don’t Trust, Always Verify

In these two future models, we have a very different spectrum of trust required, however both are unquestionably better than an exchange holding and securing customer funds. 

In this blog I will not talk about third-party custodians in much detail, as I want to focus on the cutting-edge of the spectrum: how a self-custody integration with a centralized exchange can work. 

To do this, let’s think about how an exchange manages its user’s assets. My assumption here is that the exchange is relatively prudent and employs the minimum bar of common sense in the way they secure clients assets:

Now imagine that the Deposit Addresses were replaced by self-custody wallets (e.g. a hardware wallet). Sounds great, right? However, before we get too excited, this model simply won't work for the majority of exchange activity.

Yes, you could theoretically buy some bitcoin and hook up your wallet, and the exchange could send that to your designated wallet address.But in reality, exchanges also need the ability to take funds on demand when they fall due and payable. For example...

  • Your position is liquidated? The exchange takes what is due from your wallet. 

  • You want to provide spot liquidity? The exchange knows you are good for it, because it has your funds. 

  • Want to borrow against some crypto collateral because you fancy margin trading? That’s fine, as the exchange will lend against your assets it already controls. 

So, given these challenges, how do we make self-custody work with centralized venues?

Making Self-Custody Work 

An exchange will always need an element of control over your funds. After all, your actions on an exchange have consequences. If you open a trade, you need to settle it; if you have open exposure, it should be closed out; where you want to borrow, your collateral should be accessible to the exchange when the loan falls due and payable.

The good news is that we’ve made this possible at Qredo. 

Qredo Network is a custody network, comprised of:

  1. The Qredochain, our Layer 2 blockchain, which forms an immutable record of assets and activity in Qredo Wallets, and;

  2. decentralized multi-party computation (dMPC), which signs transactions and creates wallet addresses without ever recreating the private key, and governs any withdrawals from Qredo Wallets and any usage of the Layer 2 for settlement purposes. 

Direct Debits, on a Layer 2

If an exchange permits a user to connect their Qredo Wallet to the exchange, the user has peace of mind that at all times their funds remain in their self-custody wallet. They can prefund their exchange activity without the funds having to leave their wallet — that’s the beauty of a Layer 2 integrated wallet!

Yet at the same time, where the exchange needs to take possession of an asset (e.g. to settle a liquidated trade), it can be granted an on-chain right to debit the user’s Qredo Wallet. This would be done completely transparently on-chain, giving both parties an immutable record of their activity:

We now have a way to make this work; to bring self-custody to the exchange space, or to any centralized service provider. But to make our voices heard, we need to demand it.

Join us in this battle and help spread the word of Qredo self-custodial integrations with the hashtag:


Together we can, and should, expect more from the industry!

— Josh

Ready to experience decentralized custody?

Open a Qredo Wallet