Deploy AgentOS

Templates and infrastructure for running AgentOS in your cloud.

An agent in a notebook is an experiment — to make it a product, you need to deploy it where your users are. These templates handle the infrastructure so you can focus on building agents. For runtime features, see AgentOS.

Templates

Templates handle the infrastructure. You build the agents. Pick a platform, deploy, and connect to the control plane.

Each template deploys the same core stack:

• AgentOS server — Runs agents, handles API requests (container in your cloud)
• PostgreSQL — Stores sessions, memory, traces (your database)
• pgvector — Stores knowledge for RAG (PostgreSQL extension)
• Webhook endpoints — Receives events from Slack/Telegram/WhatsApp

Same stack. Three ways to run it.

Platform	Best For	You Manage	Cost
Docker	Local dev, self-hosted	Everything	Free
Railway	Fast cloud deployment	Just code	~$20/mo
AWS	Enterprise, full control	Infrastructure-as-code	~$75/mo

Want a head start? Check out pre-built solutions like Dash, Scout, and Coda with self-learning and context retrieval already wired up.

Your data stays in your infrastructure. The control plane connects directly from your browser.

After Deployment

That’s it — your agent platform is running.

1. Verify

   Check your deployment is healthy.

2. Connect

   Open os.agno.com — one UI for your agentic software. Chat with agents, trace runs, manage knowledge, and approve sensitive actions.

3. Secure

   Configure authentication and RBAC for production.

Next Steps

Connect Interfaces

Add Slack, Telegram, or WhatsApp.

Browse Apps

Ready-to-deploy agents, teams, and workflows.

AgentOS Features

Sessions, memory, scheduling, and more.