Donation Attack

A classic example where a user deposited a large amount of CRV tokens into the Convex Finance protocol to gain disproportionate voting power over Curve's liquidity gauge weights. This allowed the attacker to direct massive CRV emissions (inflationary rewards) to a specific pool they were heavily invested in, creating an unfair advantage. The attack demonstrated how vote delegation and liquidity mining incentives can be gamed.

Attackers used flash loans to temporarily borrow a massive number of UNI tokens to submit or sway governance proposals. While the tokens were returned within the same transaction, they created a fleeting but valid voting snapshot. This exposed the need for mechanisms like vote delay, quorum thresholds, and time-weighted voting to prevent short-term capital from hijacking long-term decisions.

Protocols like Compound and Aave, where governance tokens are earned by users, are inherently susceptible. An attacker could:

• Borrow large sums of the governance token.
• Use it to vote for a malicious proposal (e.g., draining the treasury).
• The risk is mitigated by timelocks on executed proposals and guardian multisigs, but the theoretical attack vector remains a key design consideration.

Donation attacks intersect with MEV when block builders or validators can manipulate the inclusion and ordering of governance transactions. A validator with proposal-voting rights could censor opposing votes or front-run a governance outcome for profit. This highlights the need for commit-reveal schemes and resistance to censorship in on-chain governance.

Protocols have implemented several countermeasures:

• Vote Quorums: Minimum participation required for a vote to pass.
• Proposal Thresholds: Minimum token ownership to submit a proposal.
• Timelocks: Delay between a vote passing and execution, allowing for community reaction.
• Conviction Voting: Voting power increases the longer tokens are locked on a proposal.
• Whale Caps: Limiting the voting power of any single address.

The attack exploits the constant product formula x * y = k used by Automated Market Makers (AMMs) like Uniswap. By donating a large quantity of token A to the pool, the attacker drastically increases its supply relative to token B, artificially inflating B's price. This allows the attacker to then swap a small amount of the now-overvalued token B for a large amount of another asset from the same pool or a connected protocol, profiting from the manipulated price before it corrects.

In December 2020, Warp Finance lost nearly $8 million in a donation attack. The attacker:

• Donated large amounts of DAI to a Uniswap LP token (LP-DAI) pool, artificially inflating the LP token's value.
• Used this overvalued LP token as collateral to borrow other assets from Warp's lending protocol.
• The attack was possible because the protocol's price oracle used the manipulated spot price from the Uniswap pool without safeguards.

The most critical defense is using a time-weighted average price (TWAP) oracle instead of relying on spot prices. A TWAP oracle queries the price from an AMM over a time window (e.g., 30 minutes), smoothing out short-term price manipulations. Protocols like Chainlink also provide robust, decentralized oracle networks that aggregate data from multiple sources, making donation attacks economically unfeasible.

Smart contracts can implement internal checks to resist donation attacks:

• Collateralization Ratio Limits: Enforcing high collateral requirements for volatile or manipulable assets.
• Maximum Borrow Limits: Capping how much can be borrowed against a specific collateral type.
• Circuit Breakers: Pausing borrow/liquidate functions if oracle-reported prices move beyond a sane threshold within a single block.
• Using LP Token Total Supply: For LP tokens, valuing them based on the underlying reserves (sqrt(x*y)) rather than the spot price.

Donation attacks are often executed in conjunction with flash loans. An attacker borrows a massive amount of capital with no upfront collateral, uses it to execute the donation and subsequent profitable swap within a single transaction, and repays the loan. This dramatically lowers the capital barrier for the attack, making it a severe threat to under-secured protocols.

Security audits must specifically test for oracle manipulation vectors. Key areas include:

• Verifying all price feeds use TWAPs or decentralized oracles.
• Stress-testing collateral valuation under extreme pool imbalance.
• Monitoring for abnormal donation transactions or sudden, large imbalances in pools connected to DeFi protocols. Real-time analytics tools are essential for early detection.

An uncollateralized loan that must be borrowed and repaid within a single blockchain transaction. This tool is often the enabling mechanism for donation attacks, as it provides the attacker with the massive, temporary capital needed to manipulate on-chain metrics like Total Value Locked (TVL) or governance voting power without any upfront cost.

• Key Feature: Atomic execution (all-or-nothing).
• Attack Vector: Used to deposit funds, trigger a protocol's reward logic, and withdraw capital instantly.

An attack where an entity acquires enough voting power (often via token ownership) to pass malicious proposals that drain treasury funds or change protocol parameters. A donation attack can be a precursor, where an attacker temporarily inflates their governance token balance through a flash-loan-funded donation to gain voting power for a single proposal cycle.

• Objective: Control the protocol's decision-making.
• Method: Often involves manipulating token balances or voting weight.

A broad category of attacks that manipulate a protocol's financial incentives and tokenomics for profit, rather than targeting a smart contract bug. A donation attack is a pure economic exploit; it exploits the intended reward distribution logic by artificially meeting the conditions for a payout at a net profit.

• Distinction: Targets economic design, not code vulnerabilities.
• Examples: Includes inflation attacks, fee manipulation, and donation attacks.

A key metric measuring the total capital deposited in a DeFi protocol. Donation attacks often artificially inflate TVL by using flash loans to make large, temporary deposits. This can mislead users about the protocol's actual health and liquidity, and is sometimes done to qualify for liquidity mining rewards or improve ranking on analytics sites.

• Manipulation Risk: TVL is not a perfect security or health indicator.
• Impact: Can affect user perception and protocol rankings.

Incentive programs that reward users with tokens for providing liquidity. Donation attacks frequently target these programs by exploiting the reward calculation formula. An attacker donates assets to a pool to claim a disproportionately large share of the emission rewards for that epoch, often selling the rewards immediately for a profit.

• Vulnerability: Rewards based on snapshot balances or simple averages.
• Defense: Time-weighted average balances (TWAP) or minimum deposit durations.

An attack where a single entity creates many fake identities (Sybils) to subvert a system's reputation or voting mechanism. While distinct, it shares a strategic similarity with donation attacks: both involve creating a false signal of organic activity or stake. A donation attack creates a false signal of capital commitment, while a Sybil attack creates false identities.

• Common Goal: Manipulate decentralized governance or reward systems.
• Prevention: Proof-of-personhood, stake-weighting, and reputation systems.