Disagreements
860
disagreements and counting
Every PR AntFleet reviews is read by two frontier models in parallel. Only the findings both agree on get posted. The findings they don't agree on — solo flags, conflicting severity calls, classification mismatches — sit in our database, normally invisible. They're here. We don't know which side is right; the unanimous gate didn't fire. Decide for yourself.
See how AntFleet classifies reviewer conflicts in the disagreement methodology.
Older disagreements
showing 50 of 860- solo Opusdetail →Tweet ↗
CloneERC20.mintInflation drops sub-year residual when no full year has elapsed since lastMintTimestamp, but updates `currentYearStart` only inside the loop — works correctly; however `updateMintRate` bypasses `lastMintTimestamp` update when no inflation is accrued
repo a7cc2ed7·PR #3·1 week ago - solo Opusdetail →Tweet ↗
_computeCurrentFee reverts on underflow when startFee < endFee path is not blocked for equal endFee but lastFee already at endFee — but also computes uint24 underflow if startFee==endFee and durationSeconds==0 not pre-checked in DecayMulticurveInitializerHook._beforeSwap
repo a7cc2ed7·PR #2·1 week ago - solo Opusdetail →Tweet ↗
RehypeDopplerHookInitializer._onInitialization permits startFee == endFee but later code never sets lastFee == endFee, causing _getCurrentFee to keep returning startFee — only emits FeeUpdated when currentFee < schedule.lastFee, but for flat schedule short-circuit triggers; minor docs/maintainability concern: lastFee not advanced to endFee after full decay so storage write happens once and never enters the 'fully decayed' branch
repo a7cc2ed7·PR #2·1 week ago