It watches your own inbox, follows the rules you write in plain English, books and logs what your rules name, and replies as you on your say-so.
It lives on the inbox you already read, follows rules you write in plain English, turns mail into calendar events and logs, and replies as you only when you say so.
Every AI email tool so far has asked you to move in: a new client, a new app, a separate address you forward mail to. Fasrad's email assistant connects to the inbox you already read, whether that's Gmail, Google Workspace or any IMAP account, and your mail looks exactly the same afterward. It reads new messages without marking them read, moving them or deleting anything.
It acts, and that's what makes it an assistant. You tell it in plain English what you want handled: 'put every hotel and flight confirmation on my calendar,' 'log my Amazon orders to a list,' 'ask me before you reply to anyone at a client domain.' Those become standing rules it applies to every message that lands. A booking email turns into a calendar event with the address and confirmation number. A receipt turns into a row you can total up later. A reschedule you flagged turns into a one-tap question.
A morning it has already handled by the time you sit down:
It also keeps a private, searchable copy of your mailbox. Ask 'do I have a hotel booked in Munich?' or 'what did that contractor quote me last spring?' and you get a real answer in seconds, found by meaning, even if you've forgotten the exact words.
When it does reply, it sends from your own address, with your signature, only after you approve that message.
Buyers usually compare these adjacent jobs before choosing an AI agent.
Your own. It connects to the inbox you already use (Gmail, Google Workspace, or any IMAP account) with an app password. It manages the mail you actually get, so there's no second address to forward everything to.
No. It reads your mail read-only. It never marks messages read, never moves or deletes them, and files something only when a rule you wrote says to. Your inbox looks exactly as you left it.
Yes, it sends as you, from your own address with your signature, after you approve that specific message with a tap. You can also tell it to only ever draft for certain people ('never auto-send to clients'). Every send waits for your approval.
In plain English, the way you'd brief an assistant: 'calendar my hotel bookings,' 'log receipts to a list,' 'ask before replying to my boss.' There's no rules engine to build. You can edit or remove any instruction later, and they live alongside your notes.
Yes, once you turn on the searchable copy. It keeps a private index of your mailbox up to ten years back and searches it by meaning, so 'the Munich hotel confirmation' finds the right message even when those aren't the words in it. Ask in chat or from the search bar.
Time-critical mail reaches you right away: a same-day change, a cancellation, something that looks like fraud. Routine notifications wait for morning. Sensitive messages stay out of proactive alerts entirely, and when in doubt it holds them back.
Start with a 7-day free trial. Full access, no card needed. Then $49 per month. That's the whole assistant, with inbox, calendar, contacts, search and the rules included. It's in public beta and takes about 4 minutes to connect.
Agents that clear the admin. They triage the inbox, follow up after meetings, log expenses and review documents.