The mailbox that protects
your agent.
Your agent gets its own address and runs its own email. Molted screens what comes in and stops its mistakes going out.
- Claude Code
- Cursor
- Codex
- MCP
- CLI
- REST
Email is where agents make mistakes in public.
A bug in a chat stays in the chat. A bug in email lands in your customers' inboxes, from your domain, with your name on it. Molted sits between your agents and everyone they email.
Strangers can't give your agent orders.
Anyone can email your agent, and some will try to instruct it. Molted screens incoming mail, and when a message tries to take over, your agent only learns that it was quarantined.
No double sends.
Agents retry and forget what they already sent. Molted remembers, and blocks the same email to the same person twice within minutes, with a reason your agent can act on.
Hard conversations come to you.
When your agent decides a person should take over, like a refund it shouldn't grant, it hands you the thread with a note. It stays out until you hand it back.
support-bot handed you Tom Becker: a refund over $50.
Want a say? Dial it back.
Autonomy is a setting on each mailbox. Turn it down and first emails to new contacts wait for your approval, or every email does. Your agent can ask, but it can't approve for itself.
sales-outreach wants to email Jane Park for the first time.
Set up with one prompt.
- ✓Paste one prompt into your agent.
- ✓It sets up its own inbox.
- ✓It asks you only for what it can't do.
Your agent gets answers, not just a send button.
Every call tells your agent what happened and what to do next. Over MCP, a CLI or a REST API.
It knows before it sends.
Your agent can ask first: would this email go out, and if not, until when.
Mail arrives sorted, as data.
Every message comes labelled with what it is and what to do, and anything trying to steer your agent is quarantined.
It can't raise its own limits.
Agents may tighten a rule, never loosen one. That takes a person.
Priced by mailbox, not by email.
Every plan gets the whole product. Pick by how many mailboxes you run.
Free
$0
free for people
For people trying it: a form inbox, alerts, a waitlist.
- 3 mailboxes on a shared domain
- 3,000 sends a month
- 1,000 inbound emails a month
- No humanizer rewrites
Solo
$19 /mo
$1.90 per mailbox
For indie developers and freelancers, on your own domain.
- 10 mailboxes
- 3 custom domains
- 10,000 sends a month
- 5,000 inbound emails a month
- 50 humanizer rewrites a month
Team
$79 /mo
$1.58 per mailbox
For teams with agents in production.
- 50 mailboxes
- 10 custom domains
- 50,000 sends a month
- 25,000 inbound emails a month
- 300 humanizer rewrites a month
On every plan
- Prompt-injection screening
- Quarantine
- Canary tokens
- Thread checks
- Policy on every send
- Autonomy you set
- Audit log
- Keys scoped to mailboxes
- Sends to your team don't count
- Spam never counts
What each one does, and every limit side by side: pricing.
Free is for people and needs no card. An agent that signs up on its own can't use Free: it sends you a checkout link for Solo or Team, and nothing is charged until you complete it. Solo and Team start with a 3-day trial, or skip it and start right away.
Past your allowance on Solo and Team: $1.50 per 1,000 sends, $0.50 per 1,000 inbound. More mailboxes or volume? Talk to us.
Questions, answered.
Anything else? Talk to us or read the docs.
How is this different from giving my agent a Gmail account?
Do I need a Resend, Postmark or SES account?
Which agents does it work with?
What stops my agent from doing something it shouldn't?
Can I review emails before they go out?
Can a person take over a conversation?
How does my agent hear about new mail?
Can several agents share a mailbox?
How is my data handled?
Can my agent sign up by itself?
Hand it to your agent.
Set up in the time it takes to paste a prompt.