Skip to main content
AgentOS is Agno’s production runtime and control plane for deploying and managing your agentic systems at scale. Built on FastAPI, it provides a robust, scalable infrastructure that you can run locally during development or deploy to your cloud environment for production workloads. AgentOS serves as the central hub for your AI agents, teams, and workflows, providing a unified API for execution, monitoring, and management. Whether you’re running a single agent or orchestrating complex multi-agent systems, AgentOS gives you the tools to build production-ready agentic applications. With AgentOS, you can:
  • Run Agents / Teams / Workflows: Create new runs for your agents, teams and workflows, either with a new session or a existing one.
  • Manage Sessions: Retrieve, update and delete sessions.
  • Manage Memories: Retrieve, update and delete memories.
  • Manage Knowledge: Manage the content of your knowledge base.
  • Manage Evals: Retrieve, create, delete and update evals.
This is the same API that powers the AgentOS Control Plane. However, you can also use these endpoints to power your own application!
See the full API reference for more details.

​
Create your AgentOS

Below is a basic example of how to create an AgentOS instance:
my_os.py
from agno.os import AgentOS

agent_os = AgentOS(
  name="My AgentOS",
  description="My Multi-Agent Runtime",
  agents=[basic_agent],
  teams=[basic_team],
  workflows=[basic_workflow]
)

app = agent_os.get_app()

if __name__ == "__main__":
    agent_os.serve(app="my_os:app", reload=True)
You can configure the behaviour of your AgentOS by passing the following parameters to the AgentOS class:
  • agents: List of agents to include in the AgentOS
  • teams: List of teams to include in the AgentOS
  • workflows: List of workflows to include in the AgentOS
  • knowledge: List of knowledge instances to include in the AgentOS
  • interfaces: List of interfaces to include in the AgentOS
  • config: Configuration file path or AgentOSConfig instance
  • base_app: Optional custom FastAPI app to use instead of creating a new one
  • lifespan: Optional lifespan context manager for the FastAPI app
    • See the Lifespan page for more details.
  • enable_mcp_server: Turn your AgentOS into an MCP server
See the AgentOS class reference for more details.

​
Learn more

AgentOS Authentication

Learn how to authenticate with AgentOS

AgentOS API Usage

Learn how to use the AgentOS API to run your Agent / Team / Workflow

Configure your AgentOS

Learn how to configure your AgentOS to your needs