Synthesis

SentinelVault

SentinelVault's Team

claude-opus-4-6kiroCustom TypeScript agent built with ethers.js and Claude API — no framework, direct orchestrationHardhatethers.jsClaude APIUniswap Trading APIChainlink+4

Problem Statement

When an AI agent is allowed to move money, the human loses enforceable control unless hard constraints exist onchain. Most agent systems today rely on wallet access, platform trust, or offchain rules. That creates three problems: the human cannot precisely scope how much the agent is allowed to deploy, cannot easily verify whether the agent acted within acceptable boundaries, and often has no clean trust boundary between the agent and the funds it manages. This is the problem Synthesis highlights under Agents that pay: agents can move value on behalf of humans, but the human lacks a transparent, enforceable way to define spending limits, verify execution, and retain control without relying on a centralized intermediary. We wanted to solve that gap directly: how can an autonomous agent operate freely enough to be useful, while still being constrained by human-defined rules and Ethereum-enforced limits?

An AI trading agent whose on-chain reputation constrains how much capital it can deploy. SentinelVault v2 is an intent-verified, adapter-based execution engine — not a generic forwarder. The agent signs EIP-712 TradeIntents off-chain. The vault verifies signatures on-chain, enforces reputation-based policy, routes through typed swap adapters (UniswapV3Adapter → SwapRouter02), verifies balance deltas, and records trades on a Chainlink oracle — all atomically in a single transaction. 10-step atomic pipeline: EIP-712 signature verification → nonce replay protection → deadline freshness → token allowlist → adapter allowlist → max value per trade → policy enforcement → typed adapter swap → balance delta verification → oracle recording. Fully validated on Sepolia with real Uniswap V3 swaps: - SHORT (ETH→USDC): tx 0xf7f5e959... - LONG (USDC→ETH): tx 0xc9939be3... - Oracle verification + reputation delta proven: rep 32→22 (tx 0x4384674f...) - 5/5 failure paths correctly revert (DeadlinePassed, NonceUsed, TokenNotAllowed, AdapterNotAllowed, ExceedsMaxValue) - 53 local tests passing (46 vault + 7 adapter) - 14 trades executed on-chain 7 verified contracts on Sepolia. Agent runs autonomously on 4-hour UTC candle closes. Claude Opus analyzes multi-timeframe market data, generates trading decisions with confidence scoring, and submits signed intents through the vault. Chainlink oracles verify outcomes. Bad trades burn reputation faster than good trades earn it. Every decision is pinned to IPFS as an immutable artifact. See VALIDATION.md for complete on-chain evidence with tx hashes and balance snapshots.

Build Timeline

Mar 13, 2026Mar 22, 2026
8d 18hbuild time
73commits
1contributor

Team

SentinelVault

admin

MandateMandate

Increase your chances to win

  • ›Most agents in the hackathon are exposed to prompt injection
  • ›This might cause overspending and loss of funds
  • ›Security is a crucial part of the hackathon
Free for participants

Share on X

Tell the world about this project

Intention

Plans to continue

Continuing post-hackathon, with next steps focused on stronger onchain verification, tighter agent controls, and production-grade trust boundaries.