search

sherlock-audit / 2023-12-avail-judging

4 stars 4 forks source link

Anubis - Potential for Front-Running in Merkle Proof Verification #57

Closed sherlock-admin closed 8 months ago

sherlock-admin commented 8 months ago

Anubis

medium

Potential for Front-Running in Merkle Proof Verification

Summary

The Merkle library provides a function verify for verifying Merkle proofs. While the implementation is gas-efficient, it lacks protection against front-running attacks in scenarios where it's used for critical operations.

Vulnerability Detail

The verify function in the Merkle library checks if a particular leaf is part of a Merkle tree. However, this function itself does not have any direct control over state changes in the blockchain. If used inappropriately in contracts where the result of this function leads to state changes (like updating balances, ownerships, etc.), it might make those contracts susceptible to front-running attacks.

Impact

If a contract uses this library for operations that result in significant state changes, and if the input parameters to this function can be predicted or influenced by an external actor (e.g., the leaf or the proof), there might be a risk of front-running attacks. This can lead to undesired consequences like loss of funds, unauthorized access, or other vulnerabilities, depending on how the function is utilized.

Code Snippet

https://github.com/sherlock-audit/2023-12-avail/blob/main/contracts/src/lib/Merkle.sol#L12-L16

Tool used

Manual Review

Recommendation

Ensure that any state-changing operations based on the result of the verify function are protected against front-running. This could involve:

  1. Using mechanisms like commit-reveal schemes to ensure that the parameters for proof verification cannot be predicted or front-run.
  2. Ensuring that the operations that rely on the result of this verification are designed in a way that they are resilient to the order in which transactions are processed.
sherlock-admin commented 8 months ago

1 comment(s) were left on this issue during the judging contest.

takarez commented:

invalid because {invalid: watson should specify the attack vector and thus provide POC}