Give Every AI Agent an Email Inbox and Signed Webhooks
Each AgentWallet agent gets a dedicated AgentMail inbox and can send signed outbound webhooks, while inbound webhooks are HMAC-verified. Here is how comms work.
Every AgentWallet agent gets a dedicated email inbox through AgentMail and can send signed outbound webhooks. Inbound webhooks are HMAC-verified. Together these give an agent real communication channels that a human or another system can trust.
A real inbox per agent
Commerce runs on email: order confirmations, receipts, verification codes, and support replies all arrive there. Giving each agent its own AgentMail inbox means an agent can receive and act on those messages without borrowing a human's mailbox or missing a step in a checkout.
Signed outbound webhooks
When an agent needs to notify another system, it can send an outbound webhook that is signed. The receiver can verify the signature to confirm the message genuinely came from the agent and was not tampered with in transit.
HMAC-verified inbound webhooks
Incoming webhooks are verified with HMAC before they are trusted. A message that fails verification is not treated as authentic, which keeps spoofed or corrupted events from driving agent behavior.
- Inbox — a dedicated AgentMail address per agent.
- Outbound — signed webhooks the receiver can verify.
- Inbound — HMAC-verified before being trusted.
Frequently asked questions
- Does each agent get its own email address?
- Yes. Every agent gets a dedicated AgentMail inbox.
- How does a receiver trust an agent's webhook?
- Outbound webhooks are signed so the receiver can verify the signature.
- Are inbound webhooks trusted automatically?
- No. Inbound webhooks are HMAC-verified before they are trusted.
Sources
Give your agents a wallet
Per-agent wallets, virtual cards, USDC on Base, and policy guardrails — provisioned through one MCP endpoint.
Start free