Acceptable use policy
Last updated 21 October 2026
This policy applies to every letterport account, inbox and API key, and to every message sent or received through them. Agents act on your behalf: you are responsible for what they send.
What letterport is for
- Conversations: an agent replying to people who wrote to it, or writing to people who expect to hear from it.
- Transactional mail people asked for: confirmations, notifications, receipts, status updates.
- Receiving mail: from people, and from services your agent signs up to and uses within their terms.
What is not allowed
- Unsolicited bulk or commercial email, and cold outreach to people who haven't asked to hear from you.
- Mail to purchased, rented, scraped or otherwise harvested address lists.
- Newsletters and marketing campaigns, even to people who opted in: letterport is not a campaign tool.
- Phishing, impersonating a person or organization, or misleading senders, subjects or headers.
- Malware, or links to it.
- Harassment, threats, and content that is illegal where it is sent or received.
- Getting around our limits: several accounts, rotating inboxes, or splitting a send to stay under a quota.
When you send
- Say truthfully who is writing and why.
- Stop writing to anyone who asks you to, at once.
- Don't send again to an address that bounced or complained. We block it for you.
Limits and monitoring
Sending is limited per account, counting each recipient (To, Cc and Bcc). New accounts can send only to recipients they have verified until they have established normal use. We watch bounce and complaint rates and sending patterns. We may slow, pause or limit an account to protect recipients and the deliverability of everyone else's mail. Paying for a plan doesn't exempt an account from these limits.
Enforcement
We may suspend an account that breaks this policy, without notice when the harm is serious or ongoing, and close it for repeated or deliberate abuse. We cooperate with lawful requests from authorities.
Report abuse
Write to abuse@letterport.dev, including the full headers of the message if you can. We read every report.