Getting Started with Agno: Quick Start Tutorial for Building AI Agents

Getting Started with Agno: A Quick Start Tutorial

This tutorial walks you through setting up Agno and building your first AI agent. Agno is a Python SDK requiring Python 3.10+ and pip.

Prerequisites

You need Python 3.10 or later and basic familiarity with Python and command-line tools.

python3 --version

Installation

Installing Agno is a single pip command:

pip install -U agno

This installs the core SDK. For integrations (OpenAI, Anthropic, Ollama), install as needed.

Build Your First Agent in 20 Lines

Here is a complete web search agent using Agno with DuckDuckGo:

from agno.agent import Agent
from agno.tools.duckduckgo import DuckDuckGoTools

agent = Agent(
    name="Web Search Agent",
    tools=[DuckDuckGoTools()],
    show_tool_calls=True,
    markdown=True
)

agent.print_response(
    "What are the latest developments in AI agent frameworks?",
    stream=True
)

The agent uses DuckDuckGo to search the web and returns formatted markdown. With show_tool_calls=True, you see which tools the agent invokes — great for debugging.

Building a Multi-Tool Agent

Agno enables combining multiple tools. This example adds Python execution to web search:

from agno.agent import Agent
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.python import PythonTools

research_agent = Agent(
    name="Research Agent",
    tools=[DuckDuckGoTools(), PythonTools()],
    show_tool_calls=True,
    markdown=True
)

research_agent.print_response(
    "Research the population of the top 5 largest cities and calculate their average.",
    stream=True
)

Serving Your Agent as an API

Agno turns any agent into a production API endpoint with 50+ endpoints:

from agno.agent import Agent
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.runtime.api import serve_agent

agent = Agent(
    name="Search API Agent",
    tools=[DuckDuckGoTools()],
    markdown=True
)

serve_agent(agent, host="0.0.0.0", port=8080)

This starts a fully-featured API server with endpoints for chat, streaming, session management, OpenTelemetry tracing, storage, and RBAC.

Next Steps

Explore the Agno documentation to learn about context providers, human approval workflows, storage backends, and the full 100+ integration toolkit. Check out pre-built example agents like Coda, Dash, and Scout.

Conclusion

Agno makes building and deploying AI agents remarkably simple. With its clean API, extensive integrations, and production-ready features, you can go from zero to a deployed agent platform in minutes — all on your own infrastructure.