AWS Strands

Architecture Overview

Strands uses a simple Agent abstraction that wraps a model provider and a set of tools. The agent loop sends the user message plus available tool definitions to the model, which returns either a text response or tool call requests. The SDK handles tool execution and feeds results back to the model. This model-driven loop continues until the model produces a final response. Tool definitions are Python functions decorated with @tool.

When to Use AWS Strands

• AWS-native AI agent applications
• Amazon Bedrock-powered workflows
• Cloud infrastructure automation agents
• Enterprise data processing pipelines
• Multi-service AWS orchestration

Strengths & Weaknesses

Quick Start

from strands import Agent
from strands.tools import tool

@tool
def get_weather(location: str) -> str:
    """Get the current weather for a location.

    Args:
        location: The city name to get weather for.
    """
    return f"The weather in {location} is sunny, 75°F."

# Create an agent with Amazon Bedrock (default)
agent = Agent(
    tools=[get_weather],
    system_prompt="You are a helpful weather assistant.",
)

# Run the agent
response = agent("What's the weather in Seattle?")
print(response)

Features at a Glance

Notable Users

Resources