Chapter 1: Getting Started
April 13, 2026 · View on GitHub
Welcome to Chapter 1: Getting Started. In this part of Strands Agents Tutorial: Model-Driven Agent Systems with Native MCP Support, you will build an intuitive mental model first, then move into concrete implementation details and practical production tradeoffs.
This chapter gets a first Strands agent running with minimal setup.
Learning Goals
- install Strands SDK and companion tools package
- run a first tool-enabled agent call
- establish a clean local development loop
- avoid common setup mistakes
Quick Setup Pattern
python -m venv .venv
source .venv/bin/activate
pip install strands-agents strands-agents-tools
Minimal usage:
from strands import Agent
from strands_tools import calculator
agent = Agent(tools=[calculator])
agent("What is the square root of 1764?")
Source References
Summary
You now have Strands installed with a working first invocation.
Next: Chapter 2: Agent Loop and Model-Driven Architecture
Source Code Walkthrough
Use the following upstream sources to verify getting started details while reading this chapter:
src/strands/agent/agent.py— the primaryAgentclass that developers instantiate to run agent loops; this is the entry point for any Strands agent and the first class to understand when getting started.examples/— the official examples directory with minimal working agents demonstrating tool registration, model selection, and basic invocation patterns.
Suggested trace strategy:
- read the
Agent.__init__signature to understand required and optional parameters (model, tools, system_prompt) - trace a simple
agent("hello")call throughagent.pyto see the full invocation path from user input to model response - check
examples/for the simplest possible working example to use as a baseline for first runs
How These Components Connect
flowchart LR
A[agent = Agent(model, tools)] --> B[agent.py Agent class]
B --> C[Agent loop invoked with user input]
C --> D[Model response returned]