Super Safe Whitepaper — Mean Fi
Multi-sig, MFA and Treasury management for teams, projects, and DAOs on Solana
This paper discusses Super Safe (a social engineering solution) to address the security trilemma of crypto assets, through the use of BLS Threshold Multisigs, Smart Contracts and MFA (multi-factor authentication) mechanisms.
The current state of self-custody asset management with institutional-grade security is in its infancy across crypto. As the overall market cap of crypto continues to expand, the need for a highly secure, simple, private, and affordable self-custody solution remains elusive for crypto asset management.
We’ll deconstruct in this paper a solution through Mean Multisig that will address these issues while maintaining the three basic principles that underpin its development:
- 1.Security: We handle digital assets. Security and auditing are first-class citizens here.
- 2.Usability: Batteries included; it just works; as easy as apple pie; no manual needed.
- 3.Privacy: No authority or control is set over our users. Self-custody is the standard.
- Good Security + Usability at the cost of Privacy → Coinbase, Kraken, Celsius (all custody solutions)
- Good Security + Privacy at the cost of Usability → All hardware wallets (Trezor, Ledger, etc.), and all web institutional web wallets like DefiEye, Gnosis, Coinbase Institutional Custody, etc.
- Good Usability + Privacy at the cost of Security → Metamask Official , Phantom Wallet , and most popular software wallets
Crypto custody solutions in the market today always compromise by meeting only two of these three basic tenets. For example:
The primary cause for this widespread conundrum is that security-conscious wallets have come up with solutions to the security challenge from an engineering and cryptographic standpoint. Complex solutions like MPC (Multi-Party Computation) and CMP (Certificate Management Protocol) have been implemented all the way down to the chipset level. While these solutions do a lot of good from the security standpoint, their implementation is complex and creates often misunderstood processes and usability issues for the vast majority of users.
With these in mind, the aim is to deliver the next generation of self-custody asset management solutions with institutional-grade security and support for real-time finance and make it accessible and affordable for anyone to use.
Crypto asset management requires the safeguarding of a PRIVATE KEY. Whoever holds the key, holds the asset.
A private key looks like this: 5Kb8kLf9zgWQnogidDA76MzPL6TsZZY36hWXMssSzNydYXYB9KF… which, clearly, is a nightmare to remember. So we, humans, decided is better to “save” it in a better place other than our brains.
You can save it on a piece of paper, and then type that in every time you want to access your funds or make a transfer… but that’s too much typing, so you decide to send yourself an email with it in the subject line… but having to search for it every time is too time-consuming, so you save it in a Word doc with the name “BTC private key” on your desktop.
These (really unsafe methods) were the state of crypto asset management, and therefore “wallets” for the first few years of crypto (which was no bueno, if the tone didn’t give it away).
Then the great “mnemonic phrase” came to the rescue with the BIP39 standard and its cousins. It’s basically a bunch of words from a list of 2,048 words with certain entropy characteristics that make it statistically impossible for a random selection of them to be in the exact same order twice in the history of the universe. And so, a group of cyberpunks decided that instead of remembering this: 5Kb8kLf9zgWQnogidDA76MzPL6TsZZY36hWXMssSzNydYXYB9KF … we humans are going to be sooo much better remembering this:
involve layer staff express urban catch group congress addict behind drama reopen
I want some of what they are having, please! The idea is that these words are used as the input to a function that spits out the private key that was hard to remember in the first place, so… problem solved? 🤦
We crypto-geniuses basically moved the problem of “private key storage” from (1) our heads to (2) the paper to (3) a digital file, to (4) a digital file + a digital program… way to go! Some people call this progress.
Ohhh software… the solution to all our problems. Welcome to the crypto wallet orgy.
Generally speaking, a crypto wallet refers to a piece of software (usually a browser extension or a mobile app) that basically knows how to store your private key and corresponding seed phrase in the local environment where the app runs (the browser, or the OS of the device where it runs, like macOS, Windows, iOS or Android).
- Passphrase <-> Private Key conversion logic, plus
- Local storage of your crypto keys.
Some are better looking than others, some have extensibility features, and some let you do basic functions like token swaps, etc, but the essence of most wallets is the same: a self-custody solution for your crypto assets.
Then some actual PMs and Designers saw the crypto-gurus struggling and came to the rescue with some clever solutions. “What if we give you a good-old username and password and WE store your complicated private key in our servers, we’ll make it simple and awesome and familiar to you dear user, and you’ll never have to worry about your private key ever again. We’ll become your Custodian”, they said. Brilliant… everyone going bananas over Blue Labels and Caviar.
The problem these new geniuses created was that they took risks that used to be spread across thousands of individual wallets and concentrated them into single points of attack. Now the hackers don’t need to find individual users, they can simply focus on going after these servers where companies like MtGox, Cryptopia, and Coinbase (all hacked) store thousands of private keys and hold billions of assets because they are an easier target. So now these companies need to invest millions of dollars in security, violate your privacy and charge exorbitant fees for their services, and share your information with governments and marketing partners… so, bye-bye privacy there.
With centralized custody solutions, you risk PRIVACY and SECURITY for the CONVENIENCE of not having to deal with private keys. There is a popular saying in crypto that goes like this:
So, going back to owning your own keys… what’s next… well there’s this thing in security called multifactor authentication… let’s use it here. First, let’s break it down:
1.Single-Factor Authentication = Something you KNOW (like a password)
2.Two-Factor Authentication (2FA) = Something you KNOW + Something you HAVE (like a phone)
Higher factors = higher security. Ok, let’s do this crypto fam! → Welcome to hardware wallets. With a hardware wallet, we get to 2FA really quickly. It’s simple, here’s how they work:
3.Multi-Factor Authentication (MFA) = Something you KNOW + Something you HAVE + Something you ARE (like biometrics… fingerprint, face id, etc.)
1.Buy a hardware wallet
2.Set it up with the same seed phrase with a bunch of words like we covered before, then write them down and store that in your closet. The real private key is stored in the device (that’s your 2FA)
3.Set up a PIN with tiny physical buttons (this will be your equivalent to a password)
4.Install a software wallet, follow 10 tutorials online, and connect it to the hardware wallet you just bought
5.Test it and make sure it works
6.Be miserable for the rest of your life carrying around and protecting a piece of junk that removes the joy out of your crypto experience.
7.But sleep like a baby knowing that IF YOU CAN REALLY MAKE SURE NOBODY WILL EVER COPY THAT SEED PHRASE from #2, your assets will be safe.
Not only is this solution shitty because it is cumbersome to set up and pretty much nails it as the WORST onboarding experience of any product ever made by mankind, but in reality, it only removes the digital risks to your assets and moves that risk to the physical world.
Good luck if you share that SentrySafe hunk of metal with your significant other who’ll no longer be so tomorrow. Fuck! Run, run, move your shit to a new hardware device, and do steps 1–7 again, and then remove all your tokens out of the old wallet into the new one before he/she makes a gangster move to wipe you out and… wait… it’s all gone!
You just got socially engineered out of your money by an angry ex. But how is that possible?! You had what was supposed to be the safest way to guard your crypto assets.
Ok, that angry partner was not cool. So, a technical PM, watching his friend go through this pain, thought of a solution to fix the hardware wallet problem: Enter the SMART WALLET.
Smart Wallets are wallets owned by a Smart Contract instead of a public key kept by you. This little nuance gives smart wallets two super-powers:
1. You can add all the logic and rules you want to it, like account recovery process, account locks, or transfer limits per-day/per-week/per-anything-you-want-really… just like regular people are used to in any good’ol bank.
2. You remain in control of the assets in it without the responsibility of safekeeping the private key. This is accomplished by making one of the rules in #1 be that only certain other addresses can perform certain functions with the money in it.
So, in essence, smart wallets separate the storage of the assets from the storage of the keys that can access these assets. You now can simply have your “access wallet” be a regular crypto wallet that holds nothing, and move all your assets to a “smart wallet” that can be accessed any time by that “access wallet” to perform operations on those assets.
But once more, this is a technical solution to a social engineering problem. Besides the clear benefits of having a programmable wallet with a “smart wallet”, the security features are “more secure” to the extent they are more obscure.
Security by obscurity is inversely correlated to the net worth of the assets being secured. The higher the net worth, the more interest will generate, the more malicious eyes you have in it, the less obscure it becomes, and the easier it is to exploit.
Most security hacks on wallets are NOT the result of poorly implemented wallet security, but the result of social engineering attacks on the owner of the wallet hosting the private key, like the one orchestrated by the angry partner, or some phishing email the owner fell for.
Therefore it stands to reason that social engineering security should be mitigated with a social engineering solution.
Social Engineering Algorithms
Such is the case for SSS (Shamir’s Shared Secret), BLS (Boneh-Lynn-Shacham), MPC (Multi-party computation), and Multisig solutions. Even though technically they are not the same, their motivations come from similar realizations that we need more than a single point of failure to secure assets. It is not the focus of this paper to discuss the specific implementations of each of these technologies, so we’ll limit ourselves to a handy diagram:
Enter Threshold Schemes.
All these rely on the idea of maintaining a set of public keys or parts to ultimately verify a signature. However, beyond the technical complications of these methods, their weakness is having a single point of failure for the entire system on any one of their keys/parts.
A threshold scheme is fundamentally a social construct where to perform an operation with assets, a consensus must be reached first by a minimum number of participants in the system (the threshold). Threshold Multisigs, Distributed Key Generation (DKG) and more generally Multi-Party Computation (MPC) algorithms are all forms of threshold scheme cryptography. The takeaway is that if someone loses their keys, or is away on vacation, we can still sign.
Proof of work and proof of stake blockchains are essentially massive threshold schemes asking hundreds of thousands of nodes and validators to sign when they agree with a transaction. The Bitcoin network is the most attacked network in the world and has yet to suffer from a single breach of this algorithm since its inception in 2009.
A threshold multi-sig, as the name suggests, will have multiple signers (aka owners), and a minimum number of them (the threshold) must sign in order to execute transactions. This is usually referred to as n/m multisig, where n is the threshold required out of the m owners that can sign for the multisig.
Hacking a multi-sig requires coordinated social engineering attacks that are time-sensitive and operationally complex. It is the hardest to pull off, and besides serving as a deterrent for hackers, it is impossible to accurately identify all owners at the same time and have them react in the same way to the planned attack.
Even though threshold multisigs are a massive improvement in security over their technical counterparts, it lacks the simplicity and familiarity of MFA and the flexibility of the logic of smart contracts.
- Security: Driven by the Mean Threshold Multi-sig, BLS, Smart Contract Accounts/Wallets, and MFA, hacking a Super Safe wallet becomes exponentially hard as the attacker will need to perform social engineering and technology penetrations on more than one person across multiple devices to execute a malicious transaction.
- Usability: The easiest way to think about the usability of a Super Safe is like a Business Bank account (think Brex, or whatever you consider the best banking experience you’ve ever had for a business). In this “business” setting, a set of owners/authorized personnel can configure a set of rules and policies for the Super Safe, such as expenses, limits, account locking, etc. for each of the people in the organization, therefore limiting beforehand the worst-case scenario in the case anyone gets compromised.
- Privacy: Across the different technologies leveraged for the Super Safe, privacy is at the core of its decisions. The Mean Multi-sig is 100% DeFi without any centralized infrastructure. The MFA security around it has minimal centralized infrastructure blueprint to support multi-sig MFA for the Super Safe, mainly through on-device biometric verification (FaceID, TouchID, etc.), or 3rd party authenticator apps (Google Auth, Authy, etc.), or traditional communication infrastructure (email and SMS). Super Safe keeps privacy as a core tenant with no KYC/B, no custody, and no login required.
We propose combining a specific set of these technologies, specifically Smart Contracts, BLS, Threshold Multisigs, and MFAs, to bring to life a Super Safe. With clever UX and the right technology combo, we are ticking our three core tenants for self-custody asset management:
A Super Safe also offers an incredible solution to complex asset management, like account delegations, social account recovery with guardians or witnesses, dead man’s switch workflows for asset legacy and inheritance, and more. All through the simple concept of security via social engineering afforded via multisigs.
We know you want more details, more specs about Super Safe, the formulas, the design, the architecture, more, more, more.
We are, however, the kind of team that doesn’t do well with premature promises. We prefer identifying a really big problem, proposing a solution (like we did here), and iterating on it as we build it. As we do these iterations and details about the final implementation become clearer and more concrete, they will be added here.
In this whitepaper, we covered how the importance of self-custody asset management, the state of cryptographic solutions and their complexity, the security trilemma, and other industry challenges related to crypto assets custody.
We also explored what a Super Safe is, and how it leverages BLS, Threshold Multisigs, MFA, and Smart Contracts (Programs in Solana) to deliver the next generation of self-custody asset management solutions with institutional-grade security and support for real-time finance and make it accessible and affordable for anyone to use.
Want to pay your team in crypto? Token vesting? Treasury Management? Then you have come to the right place! Get Started By Launching the Mean App -> Click Here
We offer a simple, three-step process for every single token project on Solana that you can do all from the Mean Finance Dashboard.
Head to our website at https://www.meanfi.com to learn more about how we can help your token project or DAO.
If you have any questions or need help join our Discord and message one of our mods and we will help you get started quickly.
Mean Finance is bringing financial freedom to people and businesses worldwide. Through the Mean Protocol, we provide real-time cash flows and crypto asset management solutions for businesses and institutions. Teams and organizations can easily setup their crypto corporate treasuries, issue investor’s vesting contracts and run their payroll operations with real-time payment streams.
#meanfi #meandao #defi #banking #web3 #dao #staking #liquidity #moneystreaming #payroll #treasury #bonds #safes #tokenvesting #solana
Originally published at https://medium.com on June 8, 2022.