Minimax

Minimax is a competitive DeFi game where Uniswap V4 liquidity providers battle head-to-head. Players stake their LP positions and compete based on who stays in-range longest during the battle duration or their compete based on the highest earn fee LP positions. The system features two battle modes:

• Range Battles: Win by keeping your position in-range more than your opponent
• Fee Battles: Win by earning more fees during the battle period An autonomous agent monitors all active battles and automatically settles them when time expires, determining winners based on on-chain performance metrics. Cross-chain entry is seamless - users can join battles from any EVM chain using LI.FI for optimal swap routing or Arc CCTP for native USDC bridging. The agent handles the complexity: bridge funds, swap to LP tokens, add liquidity, and enter battle - all orchestrated automatically. This creates a new competitive layer for DeFi: instead of passive yield farming, LPs can now prove their range-setting skills and earn from opponents who misjudge market volatility.

Minimax is built on Uniswap V4's hook system with a multi-layer architecture: Smart Contracts (Solidity + Foundry):

• LPBattleVaultV4: Core range battle logic with Chainlink price feeds for USD valuation
• LPFeeBattleV4: Fee-based competition tracking accumulated fees per position
• BattleVaultHook: V4 hook that tracks in-range time after each swap via afterSwap callback Autonomous Agent (TypeScript + viem):
• Monitors both vault contracts for battles ready to settle
• Executes resolveBattle() transactions with full transparency logging
• Strategy loop: Monitor → Decide → Act pattern for cross-chain entry Cross-Chain Integration:
• LI.FI SDK for optimal swap routing across 5+ EVM chains
• Arc/Circle CCTP for native USDC bridging with attestation handling
• CrossChainEntryAgent combines both: analyzes user intent, compares routes (LI.FI direct vs Arc CCTP), builds execution plans Notable Implementation:
• Agent logs every action with reasoning, inputs, outputs, and tx hashes for full auditability
• LP token split calculation based on current tick position within range
• Support for battles from any chain: Arbitrum/Polygon/Optimism