The OP Stack is the open-source software that powers OP Mainnet and, according to the Optimism repository, forms the backbone of chains like OP Mainnet and Base. If you are weighing a build of your own layer-2 chain, the first job is to understand the parts. This article follows Optimism’s protocol overview and keeps to what it says.

What kind of chain it is

An OP Stack chain is an optimistic rollup. It executes transactions on its own network, publishes their data to a parent chain, and lets anyone challenge its claims about the resulting state, rather than proving each one up front. Two chains are involved: the parent chain, usually Ethereum, stores the layer-2 transaction data and holds the contracts that settle its state, while the rollup produces blocks quickly and cheaply.

The documentation calls this a derived chain. Its canonical history is defined by data published to the parent chain, so a verifier that reads only Ethereum arrives at the same chain the sequencer built. The overview describes a standard configuration with Ethereum for data availability and permissionless fault proofs, and notes that both are configurable.

The components and who runs them

The overview is clear that the OP Stack is a set of separate services rather than a single program. Its table lists these roles:

  • op-node or kona-node: the consensus layer. It sets each block’s Layer 1 origin and deposits, and derives the chain from Layer 1. It is run by the operator (the sequencer) and by anyone who wants to verify.
  • op-reth: the execution layer. It holds the mempool, orders and executes transactions and keeps state. One runs alongside each rollup node.
  • op-batcher: posts layer-2 transaction data to the data availability layer. Without it the safe head stops advancing.
  • op-proposer: posts output roots, which are commitments to layer-2 state, to Layer 1. Without it no new withdrawals can be proven.
  • op-challenger: disputes invalid proposals and defends valid ones. Without it invalid proposals go unchallenged.
  • op-conductor: optional leader election across a sequencer cluster.
  • op-deployer: deploys the chain’s Layer 1 contracts.

The consensus and execution clients run as a pair, connected over the Engine API. Two implementations of the consensus role exist: op-node, written in Go, and kona-node, written in Rust. The documentation presents this as client diversity at that layer.

A transaction, step by step

The overview walks one transaction through a chain that uses Ethereum for data availability:

  • The execution client accepts the transaction into its mempool, where it is pending.
  • The rollup node fixes the block’s Layer 1 origin and adds any deposits due.
  • The execution client selects transactions, executes them and seals the block.
  • The rollup node relays the block to peers, and verifiers see it as unsafe.
  • op-batcher compresses blocks and posts them to the data availability layer.
  • A verifier’s rollup node derives the block from the posted data, independently of the sequencer. The block is now safe.
  • Ethereum finalizes the block holding the data.
  • op-proposer submits an output root, creating a dispute game.
  • op-challenger checks the proposal against its own synced node and counters it if it is wrong.

Steps one to five depend on the operator behaving. Step six is where that dependency ends: once the data is on the data availability layer, the block’s ordering is fixed by that layer. Steps eight and nine are a separate track for withdrawals; they do not advance the chain.

What this means if you are planning a chain

The questions that follow from the documentation are practical ones. Who will run the sequencer, the batcher and the proposer, and what happens to users if one of them stops? Do you want Ethereum for data availability, which is the standard configuration, or a different choice? Will proposing and challenging be open to anyone, or restricted? Each answer changes the trust assumptions, and the documentation marks where a different choice changes the answer.

The deposit path is worth understanding early. A user who submits a deposit to the OptimismPortal contract on Layer 1 gets it included on layer 2 by the derivation rules themselves, which the documentation describes as what makes deposits censorship-resistant.

Questions people ask

Is the OP Stack a blockchain or software?

It is software. Optimism describes it as the decentralized software stack that powers OP Mainnet and forms the backbone of chains such as Base. A chain built with it is an optimistic rollup that settles to a parent chain.

What does optimistic mean here?

The chain does not prove each state update up front. Claims about state are posted, and anyone can challenge an invalid one through a dispute game.

Do I need to run every component myself?

Not necessarily. The documentation lists each component with who runs it, which can be the operator or, for verifying, anyone. Decide your split of responsibilities before you commit to a design.

Sources

← All insights