What is Fastmail?
Fastmail is a email & calendar tool. With OSP.net, your AI agent connects to it directly so it can act on your behalf instead of just answering questions about it.
What your OSP agent can do with Fastmail
- List and read recent or unread mail
- Search the mailbox by sender, subject or text
- Send plain-text email on your behalf
How to connect Fastmail
- In Fastmail, open Settings → Privacy & Security → Integrations → App Passwords (or go straight to the app passwords page) and click New App Password. app.fastmail.com → Settings → App Passwords
- Give it a name like “OSP agent” and grant it access to Mail (IMAP/SMTP). Click Generate and copy the password Fastmail shows once. Fastmail Help: app passwords
- Paste your Fastmail address and the app password here, then Connect. Your agent connects over Fastmail's standard mail servers:
Security: An app password grants access to that mailbox over IMAP/SMTP only — not your full account. Revoke it any time from the same Fastmail app-passwords page; your account password never leaves Fastmail. (This runs the same first-party mail server as the Gmail integration, just pointed at Fastmail's hosts.)
Related email & calendar integrations
Frequently asked questions
- Can OSP.net connect to Fastmail?
- Yes. Fastmail is a native OSP.net integration — you bring your own Fastmail key or token, paste it in your dashboard, and your agent restarts live.
- What can my OSP agent do with Fastmail?
- Your agent reads, searches and sends email from a Fastmail mailbox — same toolset as the Gmail integration. Specifically: List and read recent or unread mail; Search the mailbox by sender, subject or text; Send plain-text email on your behalf.
- Is Fastmail a native integration or via the gateway?
- Fastmail is a native, baked-in integration. You connect it with your own credentials, which are stored in an encrypted vault and injected only at runtime.
- Is my Fastmail data secure with OSP.net?
- Yes. Your Fastmail credentials live in an encrypted secrets vault, are injected only at container runtime, and are never written to disk in plaintext or used to train any model. Each customer runs in a fully isolated instance.