Agent Mail for AI agents
AgentMail gives AI agents their own dedicated email inboxes so they can send, receive and act on emails. Developers use it to let agents communicate with people and services by email. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Agent Mail.
Check messages in an inbox
Read the latest messages in an agent's dedicated inbox before deciding how to respond.
Send an email from an inbox
Send a reply or new message from the agent's own inbox to a person or another service.
Create a new inbox
Provision a new dedicated email inbox for an agent to use for a specific task or workflow.
5 Agent Mail actions, graded by risk.
Every Agent Mail action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
3Look things up. Allowed by default.
- agent_mail.get_messageGet Message
- agent_mail.list_inboxesList Inboxes
- agent_mail.list_messagesList Messages
write
2Create and change things. Allow, or ask the user first.
- agent_mail.send_emailSend Email
- agent_mail.create_inboxCreate Inbox
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Agent Mail in three steps.
- 01Your users connect Agent MailThey add their Agent Mail api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “create Inbox” can wait for the user to approve.
- 03Any agent can actYour agent calls Agent Mail through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('agent_mail', { read: 'allow', write: 'ask', // create_inbox destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Agent Mail connects.
Users add their Agent Mail api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Agent Mail connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Agent Mail from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Agent Mail and Arc0, answered.
Can I use Agent Mail with Claude, ChatGPT or Cursor?
Yes. Connect Agent Mail to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Agent Mail actions you allow.
How do users connect Agent Mail?
Users add their Agent Mail api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Agent Mail actions can my agent take?
5 in total: 3 read, 2 write and 0 destructive, such as “create Inbox”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Agent Mail?
Yes. Allow read actions and deny writes in the Agent Mail policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Agent Mail too?
Yes. The same Agent Mail connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Agent Mail into your agent.
Your users connect Agent Mail once, under your brand. Your agent gets 5 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan