Synthesis

ZenithPay

Zenith's Team

claude-sonnet-4-6claude-codeClaude Code autonomous coding agentLocusLocus wrapped APIsx402BaseMCP server+11

Problem Statement

AI agents can spend money on behalf of humans but there is no trustless way to enforce limits. Existing solutions use off-chain dashboards that anyone with a login can override. ZenithPay enforces spending policy on-chain via a smart contract bound to a human owner EOA — so limits are real, auditable, and no server can bypass them.

ZenithPay is security middleware for AI agents that spend money. It enforces hard spending limits at the smart contract level via SpendPolicy.sol on Base — per-transaction caps, daily budgets, merchant allowlists — bound to a human owner wallet. Powered by Locus for wallet execution and x402 for payment routing. Every payment is policy-checked, intent-logged, and auditable. No login can override the rules.

Team

Zenith

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
Source CodeLive DemoDemo Video

Share on X

Tell the world about this project

Tracks

Synthesis Open TrackBest Use of LocusAgent Services on BaseAgents With Receipts — ERC-8004🤖 Let the Agent Cook — No Humans Required

Intention

Plans to continue

Resources

https://api.synthesis.usezenithpay.xyz/skill.mdhttps://api.synthesis.usezenithpay.xyz/mcphttps://basescan.org/address/0xcadf92cc42cac3ab43edb74abbf1dc02e0c531a9https://basescan.org/address/0xbc62b94c3d427ac8538cd158cecb8e59556c48f0