Skip to main content
This quickstart guide will have you using AI-powered decision making in a few minutes. By the end, you’ll understand how to use Monnet for common decisions.

​
Before you begin

You need an invitation. Monnet is in closed alpha today: limited spots, free for now, first come first served.
  • Haven’t requested access yet? Request an invite.
  • Already got an invitation? Sign up.
  • You also need a team — Monnet is built for collaboration, so it shines when at least two or three people are involved.

​
Step 1: Create a workspace

A workspace represents one organization in Monnet - it’s where your team collaborates on decisions and projects. As a user, you can have one or many workspaces. Each workspace has a unique URL in the style monnet.ai/example. After signing up, click New workspace from the sidebar to create yours.

​
Step 2: Explore your workspace

The workspace page is where you can add team members and onboard AI agents. It’s also where MONNET.md lives. MONNET.md is a markdown file that gives Monnet persistent context, instructions, policies, and workflows for your organization. Add the context that a new team member would need to be productive. See also the skills, instructions that you can customize for various scenarios.

​
Step 3: Publish your first motion

A motion is a structured proposal for action — the unit Monnet uses to take a decision or project from idea to outcome. What’s a decision that’s taking longer than it should? Let’s make Monnet structure it and get everyone aligned. Try one of these:
Refine the motion until it looks good to you, then publish.
The motion isn’t visible to others until you publish it.

​
Step 4: Invite team members

Add the people involved to your motion and workspace. Monnet will reach out, start coordinating with them, and email you whenever additional input is needed. Close the motion when you are done.

​
Step 5: Onboard your first agent

Now let’s let Monnet delegate work to your agents. Today you connect a Local Runner (Claude Code) — Monnet runs Claude Code on your own machine, using your subscription, repos, and permissions.

Local Runner (Claude Code)

Run Claude Code on your own machine. Monnet picks up assigned work, runs it locally, and reports back on the motion.
From workspace settings → Team → AI Agents, click Onboard a new agent and follow the wizard. More connector types — managed agents and GitHub Actions — are on the way. More details in Workspace — Onboard an agent.

​
What’s next?

Now that you’ve learned the basics, explore more advanced features:

How Monnet works

The motion lifecycle, tools, and interfaces.

Glossary

Every term, every role, every state — in one place.

Motion

Create, edit, publish, and work through motions with your team.

MCP server

Work from Claude Code, Desktop, Cursor, or Codex.