Glossary

Approval Exploit

An approval exploit is theft that abuses token approvals, the permissions users grant smart contracts to move tokens on their behalf. When you use a decentralized exchange or other DeFi app, you typically approve its contract to spend a token from your wallet; if that permission is too broad or the approved contract turns malicious or gets hacked, the tokens can be taken without any further action from you.

Two variants dominate. In the first, a scam site asks you to sign what looks like a routine approval but names the attacker's contract as the spender, often for an unlimited amount; the attacker then transfers your tokens out whenever they choose, sometimes weeks later. In the second, a legitimate protocol you approved long ago suffers a contract compromise, and the attacker uses everyone's standing approvals to pull funds from wallets that interacted with it. For example, users who had granted unlimited approvals to a bridge contract have lost tokens when that contract was later exploited, despite never visiting a scam site.

The practical defenses are limiting approvals to the amount needed, periodically reviewing and revoking old approvals with tools built for that purpose, and using separate wallets for experimentation and savings. A common misconception is that moving your seed phrase to a new device fixes a bad approval; approvals live on-chain against your address, so only an on-chain revoke transaction or moving funds to a fresh address removes the exposure.