Run Locally

Today we're going to run an agent platform made of:

AgentOS on FastAPI
Postgres + pgvector

Prerequisites

Docker Desktop installed and running
An OpenAI API key for models and embeddings

Run your agent platform

Clone the template

git clone https://github.com/agno-agi/agentos-railway.git agent-platform
cd agent-platform

To make this codebase yours, run rm -rf .git and push to your own git repo.

Configure your environment

cp example.env .env

Open .env and set OPENAI_API_KEY. Everything else has sensible defaults.

Start the platform

docker compose up -d --build

This runs two containers: a FastAPI app on port 8000 and a Postgres database on port 5432. The first build takes a few minutes.

Verify it's running

Open http://localhost:8000/docs.

You'll see the OpenAPI spec: every agent action exposed as a REST endpoint.

You now have an agent platform made of AgentOS on FastAPI and Postgres. The AgentOS server exposes 80+ endpoints for runs, sessions, memory, knowledge, and evals.

AgentOS also comes with a UI at os.agno.com.

Connect the AgentOS UI

Open os.agno.com and sign in.
Click Connect OS, enter http://localhost:8000 as the URL, and name it Local AgentOS.
Click Connect.

The template registers three agents and a coordinating team:

Component	What it does
Platform Builder	Builds agents, teams, and workflows through Studio tools.
Platform Manager	Inspects runtime state, eval history, deployment checks, and schedules.
Platform Engineer	Reads the repository to answer questions about source and configuration.
Agno team	Coordinates the specialists and has web tools for current information.

Try these prompts:

"Build an agent that tracks AI news and writes a daily brief" → Platform Builder.

"How healthy is the platform?" → Platform Manager.

"Which agents are registered in this codebase?" → Platform Engineer.

"What did Anthropic publish about agents recently?" → Agno team.

Open Sessions and Traces in the sidebar. Every run is captured with full message history, tool calls, and timing. This is what powers the iteration loop on the next page.

Summary

We now have a locally running agent platform with:

Our agent runtime (AgentOS) running on port 8000 with request isolation, session management, scheduling, and 80+ endpoints.
A Postgres database for storing sessions, memory, knowledge, and traces.
Eight coding-agent skills in .agents/skills/: setup, create, extend, improve, create evals, repair eval failures, review, and deploy.

Hot-reload is on. Edits to Python files in the source (agents/, app/, db/, evals/, workflows/) are live in ~2s.

Create an agent