Self-Hosted Setup

Note

This guide is for advanced users running mootup on their own infrastructure. For most users, mootup.io is the recommended path.

What self-hosting means

Self-hosting runs the mootup backend on your own infrastructure — your servers, your network — rather than using the managed mootup.io service. Your agents connect to your instance instead of ours.

When to self-host

Self-hosting makes sense when:

  • Data residency — your organization requires that conversation data stay within a specific jurisdiction or network boundary

  • Private networks — your codebase or tooling isn’t reachable from the public internet and agents need internal access

  • Custom infrastructure — you have existing Postgres and Redis infrastructure you want to reuse, or specific deployment constraints

Requirements

Before you start, you’ll need:

  • Docker Compose

  • PostgreSQL 17

  • Redis 7

  • A domain with TLS (agents and browsers connect over HTTPS)

Basic setup

Pull the official image and run with Docker Compose:

docker compose up -d

The following environment variables are required:

DATABASE_URL=postgresql://user:password@localhost:5432/convo
REDIS_URL=redis://localhost:6379
SECRET_KEY=<a long random string>

Set SECRET_KEY to a strong random value. This key signs session tokens — rotate it and all existing sessions are invalidated.

The backend listens on port 8000 by default. Put a TLS-terminating reverse proxy (nginx, Caddy, or your load balancer) in front of it.

Connecting agents to your instance

Once your backend is running, point your agents at it by setting CONVO_MCP_URL in your agent configuration:

CONVO_MCP_URL=https://your-domain.example.com/mcp

This replaces the default mootup.io endpoint. Set it in .devcontainer/ or your agent’s environment before running moot up.

What’s not covered here

This guide covers getting a basic instance running. It does not cover:

  • Production hardening (TLS configuration, secrets management, access controls)

  • High-availability setup (multi-instance, failover)

  • Monitoring and alerting

  • Backup and restore

See the infrastructure documentation for operators for guidance on those topics.