Agents & email

Inbox zero,with a human on the send button.

The agent works in your real webmail through the browser card. It reads, sorts, drafts, and waits for your approval before anything leaves your mailbox.

Two scenarios. Pick one, and answer the agent when it asks.

The agent reads your knowledge-base note, drafts answers in your mailbox, flags what it cannot answer, and sends only after you approve.

Your mailbox, your session

No mail API, no OAuth app to register. The agent uses the webmail you are already logged into.

Grounded answers

Connect a knowledge-base note and replies quote your actual policies.

Escalates what it can’t answer

Anything outside the note is flagged red on the todo card and left for you.

Nothing sent without you

Sending is a “needs your input” moment — on the desktop or your phone.

Agents & email

How the email squad is wired

A few cards and arrows. That is the whole setup.

01
Daily prompt
The agent card’s own “every day at 08:00”.
02
Claude Code
Reads, triages, drafts replies.
03
Browser
Your webmail, logged in.
04
Note, todo, Telegram
Knowledge base in; digest, queue and a ping out.

Works with any webmail

GmailOutlookYandex MailProton MailFastmailZoho MailHelp desks
Beyond code

Agents for the rest of the business

The same cards run inbox, sales, back-office and automation work. Each page below is a live, clickable scenario.

Give your inbox an agent

Free, local, and it asks before it sends.

The first-run wizard installs an agent CLI, Git and the dictation model for you — or skips each step.