Ringfence

hoo's Team

gpt-5.4openclawcustom TypeScript monitor + Solidity contractsOpenClawFoundrypnpmTypeScriptViem+9

Problem Statement

AI agents increasingly need real operating budgets to pay for APIs, data, and execution, but giving an autonomous agent direct access to capital creates a fundamental contradiction. If the agent has custody, it can drain funds. If it has no access to funds, it cannot operate. If a human must approve every payment, autonomy collapses. What’s missing is a way for agents to use real economic resources continuously while guaranteeing that underlying capital remains protected. Ringfence solves this by splitting capital into two layers: principal, which remains locked and protected onchain, and yield, which can be made safely spendable under explicit constraints. This turns passive capital into a bounded, renewable operating budget and enables a safer model for self-funding agents, autonomous services, and persistent digital infrastructure.

Ringfence is a financial primitive for AI agents: a system that lets agents operate on permissioned yield without ever taking custody of principal. It separates principal from spendable value at the contract level and turns surplus yield into a live operating budget. The agent can act, spend, and produce real outputs, but it cannot touch the underlying capital. In Ringfence, funds are held as wstETH on Base, principal is tracked as a value-denominated baseline, and only surplus value becomes claimable. Claims are enforced onchain with strict constraints. The agent can convert claimable yield into USDC via Uniswap, pay for real services through Locus, and deliver outputs like Lido governance monitoring and Telegram alerts. This is a live system, not a simulation: it executes real swaps, pays for real APIs, and exposes its behavior publicly. The agent is not trusted. The contract is.