Strands Agents
Open-source, model-driven SDK and ready-made harness for building AI agents in Python and TypeScript, with MCP, multi-agent patterns, memory, guardrails, tracing and evals.
Tags
Description
Runs in your own process with no hosted control plane. create_harness() / createHarness() returns a fully assembled agent with defaults for model, tools, memory, sessions and context management; the lower-level SDKs (strands-agents on PyPI, @strands-agents/sdk on npm) expose the agent loop, hooks and steering. First-class providers include Amazon Bedrock, Anthropic, OpenAI and Gemini. The monorepo also ships a strands CLI and an MCP server that teaches coding agents to build with Strands. The repo was renamed from strands-agents/sdk-python to strands-agents/harness-sdk.
Agent access
Related
- OpenAI Agents SDK: Lightweight framework from OpenAI for multi-agent workflows in Python (with a JS/TS sibling SDK).
- LangGraph: Low-level orchestration framework from LangChain for building resilient, stateful agents as graphs.
- Google Agent Development Kit (ADK): Open-source, code-first toolkit from Google for building, evaluating and deploying multi-agent systems.
- Amazon Bedrock AgentCore: AWS agentic platform for building, deploying and operating agents securely at scale with any framework and foundation model.
Sources
- https://github.com/strands-agents/harness-sdk
- https://strandsagents.com/docs/user-guide/harness/
- https://strandsagents.com/llms.txt
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)