agentskillexchange Build production agent harnesses with Strands Agents
Use Strands Agents to assemble model-agnostic Python or TypeScript agent harnesses with tools, MCP, guardrails, tracing, streaming, and provider swaps.
Prerequisites
Python 3.10+ or Node.js/TypeScript, model provider credentials, optional MCP-compatible tools
Installation
Use the upstream install or setup path that matches your environment:
- pip install strands-agents strands-agents-tools
- npm install @strands-agents/sdk
- pip install hatch
- npm ci # install from repo root
Requirements and caveats from upstream:
- This monorepo contains the Python SDK, TypeScript SDK, documentation site, and supporting packages:
- | strands-py/ | Python SDK: agent loop, model providers, tools (PyPI · releases) |
Basic usage or getting-started notes:
Deliver outcomes that work. Guardrails catch mistakes before they run. Steering handlers let agents correct themselves instead of failing silently.
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](https://strandsagents.com/docs/user-guide/quickstart...
from strands import Agent
Extracted from upstream docs: https://raw.githubusercontent.com/strands-agents/harness-sdk/HEAD/README.md