Chainlink documentation: Data Feeds

Chainlink Data Feeds are the most widely-used price oracle infrastructure in DeFi. The technical documentation describes the architecture of how smart contracts query off-chain price data through Chainlink's decentralized network. Reading the documentation directly — rather than the marketing materials or third-party descriptions — is the right way to understand how oracle systems actually function and what their trust assumptions are.

The basic mechanics. A Chainlink Data Feed is an on-chain smart contract that exposes a price for a specific asset pair (ETH/USD, BTC/USD, USDC/USD, etc.). Other smart contracts can query this price by calling functions like latestRoundData() or latestAnswer() on the data feed contract. The query returns the most recent price along with a timestamp and a round identifier.

The off-chain infrastructure that produces the price is more complex. Each Chainlink Data Feed is supported by a decentralized network of node operators (typically 7 to 31 nodes per feed, depending on the asset). Each node operator independently fetches price data from multiple external sources (centralized exchanges, market data APIs, aggregators). The nodes then submit their individual price observations to an on-chain aggregator contract. The aggregator computes a median (or other consensus function) of the submitted observations and updates the on-chain price.

The trust assumptions. The price feed is trustworthy as long as a sufficient number of node operators behave honestly. The decentralized structure makes it expensive to attack a feed (you'd need to corrupt many independent nodes simultaneously). The reliance on multiple external data sources within each node further reduces the attack surface. The aggregator's median function makes the system robust to outliers — even if some nodes submit bad data, the median converges to the correct value.

The deviation triggers. Chainlink Data Feeds don't update on every block. They update when either of two conditions is met: a configured time interval has passed since the last update (e.g., every hour), or the price has moved by more than a configured deviation threshold (e.g., 0.5% for ETH/USD). The deviation threshold is a tradeoff: tighter thresholds mean more accurate prices but higher on-chain gas costs.

What this means for protocol evaluation. When you're looking at a DeFi protocol that uses Chainlink oracles, several specific questions are worth asking:

Which feed does the protocol use? The standard feed for the asset (e.g., the canonical ETH/USD feed) or a custom feed? Custom feeds may have different node sets and deviation thresholds.

What's the deviation threshold and update interval? A protocol using a feed that updates every hour with a 1% deviation threshold may be exposed to oracle staleness during fast-moving markets.

How does the protocol handle oracle failure? Most well-designed protocols have circuit breakers that pause operations if the oracle reports anomalous data or stops updating. Less-defensive protocols can be exploited by oracle failures.

Is the protocol using multiple oracle sources? Some protocols use Chainlink as the primary oracle and a secondary source (Pyth, RedStone, etc.) for redundancy. Single-oracle dependencies are higher risk.

The broader oracle landscape. Chainlink is the dominant oracle provider but not the only one. Pyth Network (originally on Solana, now multi-chain) uses a different architecture — first-party publishers (the exchanges themselves) submit prices directly, reducing the trust layer. RedStone offers on-demand oracle data pulled at query time, which can be cheaper than continuous on-chain price updates. API3 has data providers run their own nodes. Each architecture has tradeoffs and different protocols use different oracles based on their specific needs.

This is an optional, technical read. Skim the Data Feeds documentation if you want to understand exactly what's happening when a smart contract calls latestAnswer(). The on-chain query, the aggregator contract, the node operator threshold, the deviation triggers. Useful when you're evaluating whether a particular protocol is using oracles well — which is one of the most common failure modes in DeFi.