What a smart contract actually is

·5 min read·By SSP Editorial Team
SSP Academy cover: what a smart contract actually is

What a smart contract actually is

"Smart contract" sounds like a self-enforcing legal document. It's really a small program that lives on a blockchain, and seeing it that way makes a lot of wallet prompts easier to understand.

Code and storage at an address

On Ethereum and similar chains (the EVM chains), a smart contract is two things stored together at an address:

  • Code: a compiled program, published to the chain when the contract was deployed.
  • Storage: the contract's own data, such as a token's balance list or a pool's reserves.

A contract address looks like any wallet address. A normal account (an "externally owned account") is controlled by a private key and has no code. A contract account has code and no private key at all. Nobody can "sign as" the contract; it only does what its code allows.

A token like USDC is a contract. Your USDC "balance" is a line in that contract's storage that maps your address to a number. When you send USDC, you aren't moving a coin from your wallet; you're asking the USDC contract to update its records.

Contracts only act when called

A contract never wakes up by itself. Something has to call it: a transaction from an account, or a call from another contract that was itself started by a transaction. That transaction names the contract, the function to run (for example transfer or approve) and its inputs. Every node runs the same code with the same inputs and reaches the same result.

Two consequences follow. First, a contract can only do what its code says, so the code is what you are trusting. Second, every call costs gas, paid by whoever sent the transaction. That's why you need ETH to send USDC.

Immutable or upgradeable

Deployed code can't be edited in place. But many contracts are built as a proxy: a fixed address that forwards calls to a separate implementation contract, which someone can swap out. That someone, whoever holds the upgrade rights, can change what the contract does after you've started using it. We cover how to check this, and why it matters more than "the code is verified", in Verified contracts, proxies, and the upgrade key nobody mentions.

The short version: ask who holds the upgrade key.

What "interacting with a contract" means when your wallet asks you to sign

When a website asks you to "connect" and then to approve something, your wallet is preparing a transaction that calls a contract function. The prompt is asking: do you authorise this account to call this function, with these inputs, on this contract?

Some calls are obvious, like a token transfer. Others are not. An approve call gives another contract permission to move your tokens later, without asking you again. A setApprovalForAll call does the same for an entire NFT collection. If the wallet can't show you which function is being called and with what inputs, you are signing blind. Blind signing: what you are actually approving covers what to look for.

Bitcoin script and Solana programs: different models

Bitcoin has a scripting language too, but it isn't a contract platform in the EVM sense. A Bitcoin script is a short set of spending conditions attached to coins: "these coins can be spent by whoever provides two valid signatures from these two keys", for example. There's no persistent storage and nothing to call later; the script is checked once, when the coins are spent.

That simplicity is a feature. Simple scripts are easier to reason about and leave much less room for surprises. Bitcoin's 2-of-2 multisig is one of these scripts, not a deployed program.

Solana splits the idea in two. Programs hold code and keep no user data. Accounts hold data and balances, and each account is owned by a program that alone can change it. Some accounts have no private key at all; they're addresses derived from a program (called PDAs), and only that program can authorise moving what they hold, under whatever rules its code enforces.

Where SSP fits

SSP uses all three models.

  • Bitcoin and the other UTXO coins use a standard 2-of-2 multisig script built from one key in SSP Wallet and one in SSP Key. No contract is involved.
  • EVM vaults are smart contract accounts using ERC-4337 account abstraction. Your vault address is a contract (a proxy, in fact) whose code accepts a transaction only when both keys have signed. The account's upgrade function only accepts calls the account makes to itself, so it can only run through a transaction both of your keys approve. EOA vs smart account explains what that changes day to day.
  • Solana vaults are program-controlled accounts: a PDA derived from your two public keys and SSP's Solana multisig program, which only releases funds once both members have approved. That program is upgradeable, and its upgrade authority is currently held by the SSP team. That's a real trust assumption, and we'd rather you know it.

On every chain, SSP Key decodes the transaction on your phone before you approve, so you can check the contract call against what you expected.

The honest summary

A smart contract is code plus storage at an address, and it only acts when a transaction calls it. Whether its rules can change depends on who holds its upgrade rights. "Interacting with a contract" means authorising a specific function call, so read what's being called. Bitcoin uses simple one-time spending scripts instead, and Solana separates code (programs) from data (accounts).

In SSP, your Bitcoin-family funds sit behind a plain multisig script, your EVM vault is itself a smart contract account, and your Solana vault is an account controlled by SSP's program. Each needs both of your devices to move funds, with a slightly different set of things you're trusting.

Share this article

Related articles