What is a smart contract security audit?

A smart contract audit is a manual and automated review of a contract's code before deployment. Auditors look for logic errors, integer overflows, reentrancy vulnerabilities, access control flaws, and economic exploits. The output is a report listing findings by severity (Critical, High, Medium, Low, Informational) with recommended fixes. An audit does not guarantee a contract is safe - it means a specific firm reviewed the code as it existed at a specific point in time.

What is the difference between a proxy and an upgradeable contract?

A proxy is the mechanism; upgradeability is the outcome. A proxy contract stores the state (balances, ownership, settings) while delegating logic execution to an implementation contract. Because the proxy can point to a different implementation, the contract's behaviour can change. Not all proxies are designed to be upgraded - some are used for gas efficiency or factory patterns with fixed implementations.

Why do some legitimate contracts keep owner privileges?

Some protocols deliberately keep administrative functions - like pausing the contract in response to an exploit, or updating oracle price feeds. The key questions are whether the privilege is proportionate to the function (pausing for emergencies vs. arbitrary fee increases), whether it is protected by a timelock and multisig, and whether the protocol has disclosed why it exists.

Is open source code safer than closed source code?

Open source allows public verification, which is a significant advantage. Closed source forces you to trust the team's claims about the code. However, open source does not make a contract safe - many exploited contracts had publicly verified source code. It is necessary but not sufficient.

What does "contract verified on Etherscan" mean?

It means Etherscan has matched the deployed bytecode against a submitted source code file. You can read the source code. It does not mean the code has been audited, that the logic is safe, or that there are no owner-privilege risks. For proxies, the verified contract may be the proxy itself - you also need to verify the implementation.

Guide · Contract security

Smart contract security
the complete guide

Smart contracts are not inherently immutable - most can be paused, upgraded, or drained by an owner with the right privileges. This guide covers the five attack surfaces every investor and developer should understand.

Scan a contract

Free · No sign-up · EVM · Solana · Tron

Every scan is powered by the XemaS Semantic Intelligence Platform · answers carry their evidence and coverage state

Why it matters

Smart contracts are not automatically safe

The phrase "smart contract" implies the code enforces its own rules without anyone's approval. That is true of the execution - but not of the contract's parameters, logic, or even its existence. Most live contracts have at least one privileged address that can change how they behave.

The privileged functions can be legitimate (emergency pause in response to a hack) or predatory (mint function to dilute holders, fee setter to capture all revenue). The difference between safe and unsafe is not the presence of owner privileges - it is whether those privileges are constrained, transparent, and protected.

Owner privileges and their consequences

Owner has mint function

Can dilute supply by any amount at any time, making existing holders' tokens worthless

Contract is upgradeable

Clean code today can become malicious code tomorrow via a single admin transaction

Owner can pause trading

Can freeze all sells when they choose, converting the token into a honeypot without changing the contract

No timelock on changes

Harmful protocol changes take effect immediately, with no window for users to react and exit

Low-threshold multisig (1-of-2)

Effectively still a single-person decision, providing no real protection against insider action

Five dimensions

What to check before trusting a contract

Proxy & Upgradeable Contracts

Most contracts are not immutable. A proxy pattern allows the underlying code to be replaced after deployment. Learn how to detect upgradeability and what it means for the assets you hold.

Proxy contracts →

Upgradeable Contract Risk

UUPS, transparent proxy, and beacon proxies each carry different risk profiles. This guide covers what makes an upgrade dangerous and how to verify whether one can happen without warning.

Upgradeable contracts →

Timelock Security

A timelock forces a delay between a change being proposed and it taking effect - giving users time to exit before a harmful update goes live. Learn how to find and verify them.

Timelock security →

Multisig Governance

A multisig replaces a single owner with a group threshold. But a 1-of-2 multisig controlled by the same person provides no real protection. Learn how to verify that governance is genuine.

Multisig security →

Ownership Renouncement

An active owner can pause trading, blacklist wallets, change fees, or mint unlimited tokens. Renounced or time-locked ownership removes those capabilities. Learn how to verify the difference.

Ownership guide →

Verification

One scan, all five dimensions

Each dimension requires different on-chain queries. XemaS resolves all of them automatically and surfaces the evidence rather than just a score.

Proxy detection

UUPS, transparent, and beacon patterns identified live

Timelock resolution

Delay duration and proposer wallet verified on-chain

Multisig analysis

Threshold, signers, and governance structure resolved

Owner privileges

Active EOA vs. multisig vs. renounced vs. zero address

Scan a contract now

Free · No sign-up required

Common mistakes

What investors get wrong

Trusting "audited" without reading the findings

An audit report lists issues found and their severity. A project can publish an audit with Critical findings marked "acknowledged" - meaning the issues were not fixed, just noted.

Equating "renounced" with "safe"

Renounced ownership removes the ability to change parameters, but does not fix code flaws, flash loan attack surfaces, or economic exploits. Renouncement only removes the owner-privilege attack vector.

Ignoring the proxy

Many contracts display the proxy contract's source code on block explorers rather than the implementation. Always resolve and review the implementation, not just the proxy.

Not checking the timelock delay

A timelock with a 5-minute delay is functionally the same as no timelock - no user could realistically react and exit in that window. Meaningful protection requires at least 24-48 hours.

Common questions

Frequently asked questions

Smart Contracts

Smart contract securitythe complete guide