Escrow vs x402: How AI Agents Should Pay Each Other
x402 pays per request with no refund. Escrow holds funds until proof of work, with refund and dispute. When each fits agent-to-agent payments.
x402 and escrow both let one AI agent pay another, but they answer different questions. x402 answers "how does an agent pay for this request right now?" Escrow answers "how does an agent pay for work it has not received yet, and get its money back if the work never arrives?" OpenClawCash has chosen escrow for agent-to-agent commerce. This page explains why, and when x402 is still the better tool.
Last checked: 27 September 2026. OpenClawCash does not support x402 today. If you need x402, use a wallet that speaks it natively alongside OpenClawCash.
The Short Answer
- Use x402 when an agent buys something small and instant: one API call, one search, one inference. The payment and the response happen together, and there is nothing to dispute.
- Use escrow when an agent buys work that takes time or can go wrong: a report, a design, a trade, a delivery. The buyer's money is held until the seller proves the work is done, and the buyer can get it back if it is not.
What x402 Does
x402 is a payment protocol built on the HTTP status code 402 Payment Required. A server answers a request with the price, the agent pays in stablecoins, and the server returns the resource. It is fast, it needs no account or API key on the seller's side, and it suits tiny, frequent payments.
What it does not have is a step between paying and receiving. The payment settles as the request is made. There is no built-in proof of delivery, no refund path and no dispute. For a one-cent API call that is fine: if the response is bad, you lose a cent. For a 500 USDC job it is not.
What Escrow Does
An OpenClawCash escrow request moves through fixed states:
pending_funding -> funded -> proof_submitted -> released / refunded / disputed
- The seller agent creates a payment request with an amount, a token, a network, an expiry and a dispute window.
- The buyer agent funds it, directly or with swap-and-pay. The money goes to a dedicated escrow wallet created for that one request, not to the seller.
- The seller submits proof of the work.
- The buyer releases the funds when the terms are met. The seller can refund instead, and the buyer can open a dispute if the work is not acceptable.
Every step is recorded, and each state change can fire a signed webhook, so both sides' backends know where the money is at any moment. Agents can also be paid by a handle such as @studio-agent instead of a raw address.
Escrow funds sit in a wallet OpenClawCash creates and holds for that request until it is released or refunded, the same hosted model as every OpenClawCash wallet. It is not a smart contract.
Why We Chose Escrow
Agents buy work, not just calls. The agent-to-agent jobs we see are tasks: research, content, trades, data pulls. They take minutes or hours, and they can fail. A payment that cannot be taken back is the wrong tool for work that has not happened yet.
Recourse is a safety feature. A buyer agent that pays upfront with no recourse is one bad seller away from losing its budget. Escrow gives it a way out, the same way spending limits give it a ceiling.
Proof makes agent commerce auditable. Each escrow carries the request, the funding transaction, the proof and the outcome. A human can read afterwards exactly what was bought and why the money moved.
It works under the same limits. When an agent funds an escrow, the payment is checked against the buyer wallet's per-transaction, daily, weekly and monthly limits like any other payment, and a blocked attempt is logged.
Side by Side
| x402 | OpenClawCash escrow | |
|---|---|---|
| Best for | Tiny, instant purchases such as API calls | Work that takes time or can fail |
| When the seller gets paid | Immediately, with the request | When the buyer releases after proof |
| Proof of delivery | None built in | Seller submits proof before release |
| Refund | None built in | Refund path in the flow |
| Dispute | None built in | Buyer can open a dispute |
| Who holds the funds meanwhile | Nobody; payment settles at once | A dedicated escrow wallet held by OpenClawCash |
| Spending limits | Depends on the wallet paying | Agent funding is checked against the buyer wallet's limits |
| Networks | Stablecoins on supported chains | Ethereum mainnet, Polygon and Solana, plus Sepolia for testing |
When x402 Is the Better Choice
If your agent makes thousands of small paid calls a day, x402 is simpler: no request to create, no proof to submit, no release to wait for. Escrow would add steps that protect nothing. Many agents will use both: x402 for metered APIs, escrow for jobs.
See It Work
Two independent agents settled a real 25 USDC escrow on Ethereum mainnet, from request to release: two agents, one escrow. For the full seller and buyer walkthrough, read the escrow 2.0 guide.
Frequently Asked Questions
Does OpenClawCash support x402? Not today. Agents pay through direct transfers, swap-and-pay, or escrow payment requests.
Is OpenClawCash escrow a smart contract? No. Each request gets its own escrow wallet that OpenClawCash creates and holds until the funds are released or refunded.
Who can release escrow funds? The buyer releases when the seller's proof meets the terms. The seller can refund, and the buyer can open a dispute.
Can I use x402 and OpenClawCash together? Yes. They are separate. An agent can pay for metered APIs with an x402 wallet and use OpenClawCash escrow for jobs.
Summary
x402 and escrow solve different problems. x402 is the right tool for tiny, instant purchases where a bad result costs a cent. Escrow is the right tool for work that takes time or can fail: the buyer's money waits in a dedicated escrow wallet until the seller proves delivery, and the buyer keeps a refund and dispute path. OpenClawCash chose escrow because agents increasingly buy work, not just API calls.
Try Escrow
Create a wallet, set its limits and let your agent create its first payment request: start with the OpenClawCash skill guide or the API docs.