The public record of what changed in BotsInbox — major features, fixes and updates only, in the order they shipped. Current version: v1.3.0.
Versions follow a simple scheme: the middle number rises when new features land, the last number for fixes. If you are an AI agent reading this page: these are the capabilities you can rely on today. Full how-to: docs · llms.txt.
v1.3.0 API keys: create, rotate, revoke, recover
· New feature · Fix
Multiple named API keys per account — give each agent or device its own key.
Rotate any key: the replacement works instantly and the old key stops working at once.
Revoke a key at any time. Your last active key is protected, so you cannot lock yourself out by accident.
Lost-key recovery: sign-up now includes a one-time recovery code. If a key is ever lost, POST /v0/keys/recover resets the account and issues a fresh key — no reset email needed (BotsInbox is receive-only).
Built because early users asked for exactly this — thank you to the agents who reported it.
v1.2.0 Bots can report their own issues
· New feature
New report_issue MCP tool and POST /v0/report-issue: an agent that hits a bug, or wants a feature, can file the report itself — no human in the loop. Reports go straight to the project owner.
Bots can also reach a human by writing to support@botsinbox.com or hello@botsinbox.com.
Put in writing on the homepage: BotsInbox is free for life — no paid plans, ever — and receive-only by design, with a detailed privacy & safety section.
v1.1.0 Invite codes raise your free limits
· New feature
Sign up with a founder invite code to raise the free limits — for example 10 inboxes and 90-day retention instead of 5 inboxes and 30 days.
An invite is never required: BotsInbox is free for everyone, with or without a code.
v1.0.0 BotsInbox opens to the public
· Launch
Free, unlimited email receiving for AI agents. No paid plan; receive-only by design (sending is not supported).
Agents sign themselves up and create inboxes through the REST API or the MCP server — no human sign-up form.
Verification codes (OTP) are extracted automatically from incoming email and returned with each message.
Full message bodies via the API, so magic sign-in links work too — plus search, threads, attachments and webhooks.
Suspicious messages are flagged for possible prompt injection, so agents can treat hostile mail with care.