Skip to content

/harnessStrands harness is a state-of-the-art, fully assembled agent harness./harness-sdkThe Strands Harness SDK for building an agent harness from the ground up./boxAn agent sandbox that pairs OS containment with Dogwood policy./shellA virtual shell designed for AI agents to use safely./evalsValidate your agent before you ship./labsExperimental projects pushing the boundaries of what agents can do.
CoursesGuided lessons from your first agent to multi-agent systems.ExamplesRunnable samples you can drop into your own project.
Integrations
Contribute
Blog

/harnessStrands harness is a state-of-the-art, fully assembled agent harness.

Get started
Configure the agent
Built-in tools
Take Strands harness to production
Reference
- Configuration reference
- Versioning & Support

/harness-sdkThe Strands Harness SDK for building an agent harness from the ground up.

Get started
Migrate
Build guides
Run guides
- Deploy to production
  - Operating Agents in Production
  - Amazon Bedrock AgentCore
    Overview
    Python
    TypeScript
  - AWS Lambda
  - AWS Fargate
  - AWS App Runner
  - Amazon EKS
  - Amazon EC2
  - Docker
    Overview
    Python
    TypeScript
  - Kubernetes
  - Terraform
  - Nx Plugin for AWS
- Observe your agent
- Secure for production
Components
- Agent loop
- Production lifecycle controls
- State
- Storage
  - Overview
  - Search Strategies
- Snapshots
- Prompts
- Hooks
- Hook events
- Middleware
- Context management
- Retry strategies
- Structured output
- Interrupts
  - Overview
  - Interrupts in Multi-Agent Systems
- Models
- Tools
  - Tool result format
  - MCP Transports
  - Executors
  - Vended Tools
    Overview
    Edit
    File editor
    Read
    Write
    HTTP request
    Notebook
    Bash / shell
    MCP Router
    Sleep
    Handoff to User
    Stop
    Web fetch
    A2A Client
    Python REPL
    Subagent
    Swarm
    Graph
- Memory
- Plugins
- Interventions
- Sandbox
- Multi-agent
- Streaming
- Voice
- Conversation management
- Experimental
  - AgentConfig
Reference

/boxAn agent sandbox that pairs OS containment with Dogwood policy.

Overview
- Getting started with Box
Guides
Security
- Security: what Box governs, and what you own
- Experimental native filesystem policy
Reference

/shellA virtual shell designed for AI agents to use safely.

Get started
Guides
Components
Security
- Security model
Reference
- Reference

/evalsValidate your agent before you ship.

Get started
Build guides
- Generate test cases
- Plan topic coverage
Run guides
Components
CLI

/labsExperimental projects pushing the boundaries of what agents can do.

CoursesGuided lessons from your first agent to multi-agent systems.

ExamplesRunnable samples you can drop into your own project.

TypeScript
- Learn
- Deploy
  - AgentCore
  - Next.js Vercel Adapter
- UX Demos
  - Dynamic UI

Integrations Contribute Blog

/examples

TypeScript
- Learn
- Deploy
  - AgentCore
  - Next.js Vercel Adapter
- UX Demos
  - Dynamic UI

Select theme

On this page

Overview

On this page

Overview

Streaming

View the complete sample on GitHub

Coming Soon

Previous
Model Providers Next
Agent State

Open an issue Ask on Discord

Privacy
Site terms
License
Cookies
Preferences
© 2026 Amazon Web Services, Inc. or its affiliates