Strands Agents

// Overview

Strands Agents is an open-source SDK from AWS for building and running AI agents. It takes a model-driven approach, so you define an agent and the tools it can call, and the model drives the loop that decides what to do next. The project ships both a Python SDK and a TypeScript SDK from the same monorepo.

It is aimed at developers who want to build agents that range from a simple conversational assistant to a multi-step autonomous workflow. Both SDKs default to the Amazon Bedrock model provider, but the SDK is model agnostic and also supports Anthropic, OpenAI, Gemini, Ollama, and other providers, so you can swap backends without rewriting your agent code.

As an app framework for LLM orchestration, Strands provides the building blocks you usually have to assemble yourself: an agent loop that traces each decision, hooks to intercept steps, context management, execution limits, MCP support, streaming, and structured output. That lets you focus on the agent's behavior instead of the plumbing around it.

// What it does

• Model agnostic with first-class support for Amazon Bedrock, Anthropic, OpenAI, and Gemini, plus many more and custom providers
• Built-in tool use, including a separate strands-agents-tools package and Zod-typed tools in TypeScript
• Agent loop that traces every decision by default, with hooks to log, validate, or redirect any step
• Native MCP, streaming, multi-agent patterns, and structured output
• Available as both a Python SDK (pip) and a TypeScript SDK (npm) from one monorepo
• Guardrails and steering handlers so agents can correct themselves instead of failing silently

// Getting started

Both SDKs default to the Amazon Bedrock model provider, so you'll need AWS credentials configured and model access enabled for Claude Sonnet. The Quickstart Guide covers configuring other providers such as Anthropic, OpenAI, Gemini, and Ollama.

pip install strands-agents strands-agents-tools

from strands import Agent
from strands_tools import calculator

agent = Agent(tools=[calculator])
agent("What is the square root of 1764")

npm install @strands-agents/sdk

import { Agent } from '@strands-agents/sdk'

const agent = new Agent()
const result = await agent.invoke('What is the square root of 1764?')
console.log(result)

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Build a conversational assistant that can call tools such as a calculator or custom functions
• Create multi-agent workflows where agents coordinate to complete a task
• Prototype an agent locally on one model provider, then swap to another backend for production without changing your agent code
• Connect an agent to external systems through MCP and stream its responses

// How Strands Agents compares

Strands Agents alongside other open-source agent frameworks & builders tools AI/TLDR tracks, ranked by GitHub stars.