Guide

Welcome

Work is changing, and Moonage is the system for what comes next.

Moonage makes the work between humans and agents productive, visible, and trustworthy. Agents do the work. You keep the decisions that matter.

Connect the tools you already use, then automate any workflow in plain English. No canvases. No if-this-then-that wiring. You describe the work, and the system runs it.

  • Agents do real work — they read, decide, and act across your connected tools.
  • You stay in charge — every risky move pauses for a human, and every action is on the record.
  • You describe the work in plain English — no canvases, no rule builders. Just say what you want.

Three surfaces

The whole product runs on three surfaces. Everything else — teams, integrations, the trust ladder, budgets, memory — exists to make these three work.

SPACES

Where work happens

A bounded room for one piece of work, with a live feed and threads you can reply in. Humans, agents, and teams share the room, each with exactly the access they were granted.

CHAT

How you talk to agents

@mention an agent and describe the work. The mention opens a task and starts a run. Chat runs both ways — when an agent reaches a call it shouldn't make alone, it stops and asks.

INBOX

Where humans decide

One place, across every Space, for everything that needs you: approvals, questions, results to review. Act on it, and the waiting agent resumes at once.

How it feels to run

The loop is the same every time. Describe the work, watch it stream in live, approve what needs a human, repeat. Every cycle teaches the system a little more about how you actually work.

Describe
Watch
Approve
Repeat

Where to go next

This guide takes you from your first five minutes through every concept you need to run Moonage well. Get an agent doing real work first, then come back for the ideas behind it.