Quick Answer
Stablecoin protocols are uniquely difficult to audit because their security depends not only on code correctness but on mechanism design soundness. A technically correct implementation of a flawed peg mechanism will fail — as Terra demonstrated at $40 billion scale. Security review must cover both the code and the economic logic.
Primary risks: minting/burning access control, blacklist function security, reserve attestation trust. Smart contract surface is smaller but admin key custody is the primary attack vector.
Oracle security for collateral pricing, liquidation logic correctness, stability pool reentrancy, debt ceiling governance. The most complex smart contract surface of any stablecoin type.
Reflexive peg mechanisms with seigniorage companions have demonstrated death spiral failure modes. Terra/LUNA wiped $40B. The mechanism design must be reviewed alongside the code — technically correct code implementing a flawed mechanism still fails.
Both collateral management AND algorithmic components require audit. The interaction between the two components — how the protocol shifts between full and partial collateralization — introduces additional attack surface beyond either component alone.
| Vulnerability | Severity | Description | Example |
|---|---|---|---|
| Oracle Price Feed Manipulation | Critical | Stablecoins that maintain their peg via collateral liquidations depend on accurate, manipulation-resistant price feeds. A manipulated oracle can cause either improper liquidations (when collateral is priced below true value) or allow undercollateralized positions to remain open (when collateral is priced above true value). This is the attack class that caused the bZx protocol losses ($8M, 2020) and multiple subsequent DeFi exploits. | A CDP stablecoin using a spot price Chainlink feed for collateral valuation. A flash loan inflates the collateral token's price just long enough for a borrower to take out an undercollateralized loan before the price reverts. |
| Algorithmic Peg Spiral (Death Spiral) | Critical | Algorithmic stablecoins that use a seigniorage model (burn/mint of a companion governance token to maintain peg) are vulnerable to reflexive death spirals. When the stablecoin de-pegs below $1, the system mints more governance tokens to incentivize buying — but if confidence is lost, the newly minted tokens also lose value, accelerating the peg failure. Terra/LUNA's $40B collapse in May 2022 is the canonical example. | An algorithmic stablecoin that requires 100% of its peg to be maintained by a companion token with a $200M market cap. A sudden $100M sell pressure on the stablecoin requires minting $100M of companion tokens — if the companion token market can't absorb this, both spiral to zero. |
| Liquidation Parameter Governance Attack | High | Stablecoin protocols with governance-controlled liquidation parameters (collateral ratio thresholds, liquidation penalties, interest rates) can be exploited via governance attacks. A malicious governance proposal that lowers collateral ratios or disables liquidations can allow protocol insolvency. The Beanstalk governance attack ($182M, April 2022) used a flash loan to acquire enough governance tokens to pass a malicious proposal in a single transaction. | A stablecoin governance contract with no timelock on parameter changes — a proposal to change the collateral ratio from 150% to 101% passes and executes in the same block, instantly making most positions liquidatable. |
| Collateral Redemption Reentrancy | High | Stablecoin redemption functions that release collateral to burners before updating the debt accounting are vulnerable to reentrancy. An attacker who receives control flow during the collateral release can re-enter the redemption function and claim additional collateral for the same amount of burned stablecoin. | A Frax-style stablecoin redemption function that sends ETH collateral to the redeemer via a callback and then reduces the stablecoin supply — an attacker using a smart contract wallet re-enters on the ETH receipt before the supply reduction. |
| Stale Price During Market Stress | Medium | During extreme market events, Chainlink price feeds can lag real-world prices by several blocks due to heartbeat intervals and deviation thresholds. Stablecoin protocols that liquidate based on stale prices can incorrectly trigger (or fail to trigger) liquidations, causing protocol losses or preventing healthy liquidations during exactly the moments when they're most needed. | A CDP stablecoin that uses a Chainlink feed with a 1-hour heartbeat. During a rapid 20% ETH price drop, the feed doesn't update for 45 minutes. Protocol continues issuing stablecoin against ETH priced at the stale rate while the actual collateral value has dropped below the required ratio. |