What is proof-of-execution?

Proof-of-execution is a cryptographic, timestamped record that an AI agent (or any automated system) performed a specific action — signed and anchored on a blockchain so anyone can verify it independently, forever.

Why AI agents need it

When an autonomous agent sends money, approves a contract, or files a claim, "the AI did it" is not evidence. Screenshots can be faked and logs can be edited. Proof-of-execution gives you a record that no one — not even you — can alter after the fact, which you can show a customer, an auditor, or a court.

How it works

1. Sign — the agent's output is hashed and Ed25519-signed.
2. Anchor — the fingerprint is written to multiple blockchains (XDC, Base, Hedera, XRP, Stellar).
3. Verify — anyone opens a public receipt link or calls the verify API to confirm the signature and the on-chain transaction.

Proof-of-execution vs. logging

A log lives in your database and can be quietly changed. A proof-of-execution receipt is anchored on public blockchains and signed — it is tamper-evident by design and verifiable by a third party who does not trust you.