Smart contracts are deterministic: given the same inputs, they always produce the same outputs. That's a feature — it's what makes them trustless and auditable. The limitation: they can't natively access data outside the blockchain. They can't check the price of SOL, the EUR/USD exchange rate, or tomorrow's weather.
Oracles solve this by bringing external data on-chain in a way that's reliable enough for smart contracts to act on. They're critical infrastructure — without oracles, most of DeFi breaks.
Why Oracles Matter
Consider a lending protocol: it needs to know the current price of your collateral to decide whether to liquidate you. That price comes from market data — which lives off-chain. The protocol needs an oracle to bring that price on-chain.
Consider a stablecoin: it needs to know what the USD price of ETH is to maintain its peg. Oracle.
Consider a prediction market: it needs to know the outcome of a real-world event. Oracle.
The oracle problem — how to get trustworthy external data into a trustless system — is one of the fundamental challenges in blockchain design.
Chainlink: The Dominant EVM Oracle
Chainlink is the largest decentralized oracle network. It aggregates price data from multiple independent node operators, each of which fetches data from multiple sources and signs their report. The aggregated median is posted on-chain.
How it works: Dozens of independent Chainlink nodes fetch ETH/USD prices from multiple exchanges, sign their data, and submit it. The Chainlink contract takes the median of all submissions and posts it as the authoritative price. Corrupting the feed requires corrupting a majority of independent nodes simultaneously — economically impractical.
Chainlink powers most Ethereum, Base, Avalanche, and Polygon DeFi protocols. Aave, Compound, Synthetix, and hundreds of others depend on Chainlink price feeds.
Pyth Network: Solana's Native Oracle
Pyth is the dominant price oracle on Solana — and increasingly cross-chain. It uses a "pull" model rather than Chainlink's "push" model:
Push oracle (Chainlink): Data is regularly pushed on-chain regardless of whether anyone needs it. Higher gas cost, but always available.
Pull oracle (Pyth): Price data is available off-chain (on the Pythnet appchain) and pulled on-chain only when a transaction needs it. More gas-efficient, higher update frequency (sub-second on Solana).
Pyth sources data directly from institutional market makers, exchanges, and financial data firms — not from DEX prices. This makes it more reliable for liquid assets and more resistant to DEX manipulation attacks.
Jupiter, Drift, Marginfi, and most major Solana protocols use Pyth for price feeds.
Oracle Risks
Oracles are a major attack surface:
Price manipulation — If an oracle sources price data from a single DEX, an attacker can temporarily manipulate that DEX's price (e.g., via flash loan) to trick the oracle. This has caused multiple DeFi hacks. Protocols using aggregated oracle feeds from multiple sources are more resistant.
Oracle downtime — If the oracle stops updating (network congestion, node failures), the last reported price stays on-chain and becomes stale. During extreme volatility, stale prices have caused incorrect liquidations and losses.
Centralization risk — A small oracle team with upgrade keys can change the oracle logic. Fully decentralized oracle networks distribute this risk; smaller oracle projects may not.
When evaluating a DeFi protocol, always check what oracle it uses. "Custom oracle" or "team-controlled price feed" is a significant risk flag.
Explore SovereignSwap — built on audited Solana infrastructure →