OpenRMM documentation
An open-source, self-hosted RMM. The AI assistant reads your fleet and proposes; nothing runs on a machine until a person approves it.
These guides live with the code in the OpenRMM repository: this site is rebuilt from it on every change, and every page has a link to edit its source.
Get started
Section titled “Get started”Self-hosting quick startDocker Compose, the published images and automatic HTTPS on your own server.
OverviewWhat OpenRMM is, what is built and what is intentionally left out.
Production deploymentTLS, secrets, agent installers, backups, upgrades and rollback.
Run from sourceBuild and run the server, console and agent for development.
Using OpenRMM
Section titled “Using OpenRMM”AI assistantHow the chat reads your fleet, proposes jobs, and why it can only propose.
Remote accessRemote desktop, file transfer and terminal, plus MeshCentral and RustDesk.
Monitoring, policies and alertsCompanies, sites, roles, checks, tasks, policies and alerts.
Update managementScan, approve and install OS updates in maintenance windows.
MCP serverConnect Claude, ChatGPT or another assistant with the same permissions and approvals.
Single sign-onSign in through any OpenID Connect provider.
Security and development
Section titled “Security and development”Security overviewThe guarantees enforced in code, roles, two-factor sign-in and the audit trail.
ArchitectureThe components and how a proposal, an agent and a remote session flow through them.
API referenceAuthentication, roles and the REST endpoints behind the console.
ContributingSign-off, the contributor agreement, the tests and what gets a pull request merged.