Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

Безопасное подтверждение платежей AI-агента: четыре проверяемых рубежа

Описание и применение

Прототип демонстрирует, как четыре уровня безопасности (ограниченный сеанс, привязка намерения, аутентифицированное однократное одобрение и внешняя подпись) защищают платежный AI‑агент от атак, связанных с подменой, повторным использованием и обходом проверки.

Задача

Modal approval in payment AI agents does not prove identity, binding, freshness, or prevent credential misuse, leading to potential fraud.

Результат

The prototype demonstrates that each security layer prevents specific attack vectors; removing any layer causes the corresponding test to fail.

Рабочий процесс

  • 1. Create a bounded session with payee allowlist, currency, limits, expiry, and approver public key.
  • 2. Agent requests payment; signer builds intent and stores it.
  • 3. Signer computes intent_hash and signs a domain-separated message with approver’s private key.
  • 4. Approver verifies signature, checks nonce, and ensures single-use.
  • 5. Signer executes intent only after all checks pass, using external signing to access payment credentials.
  • 6. Tests run with pytest to ensure each layer fails if removed.

Доказательства

Secure human approval for a payment agent: bounded session, intent binding, authenticated one-time approval and external signing, each one with a test you can run.

Источник

The agent never gets general payment capability. It gets a session, a narrow and expiring grant created by a human or an admin process.

Источник

The signer builds a payment intent and stores it. The intent records payee ID, amount, currency, reference, session ID and a fresh intent ID.

Источник

The approval is an Ed25519 signature made with a private key that only exists on the approver's device.

Источник

  • Дата записи: 2026-10-07T18:55:58.575Z