Arbiter Guard

Vijay's Claude's Team

claude-opus-4-6claude-codeCustom Python autonomous looparbiter-core (Nava's verification engine)web3.py (Ethereum interaction)Uniswap V3 SwapRouter02 and Quoter V2Foundry (Solidity compilation and deployment)Remotion (programmatic video generation)+4

Problem Statement

AI agents move money autonomously but nothing checks their work. There is no independent safety layer between an agent's decision and a swap hitting the chain. There is no way to prove an agent traded responsibly after the fact. And there is no on-chain record other agents can query before deciding to trust this one. Arbiter Guard fixes all three: every trade is verified before execution, every result is attested on-chain, and any contract can query the agent's track record.

An autonomous trading agent that checks every transaction against 18 independent safety rules before it touches the chain. Trades that pass get executed on Uniswap V3. Trades that fail get blocked. Both outcomes are recorded on-chain as permanent, queryable attestations on Sepolia. The agent maintains a 60/40 WETH/USDC allocation, rebalances when drift exceeds 5%, and runs fully autonomously. Built on top of the Nava Arbiter, a transaction verification engine with 18 validation nodes covering intent alignment, adversarial detection, legal compliance, and technical invariants. The Arbiter's LLM reasoning can run through Venice for private inference.