AWS Open-Sources Strands Agents SDK to Boost AI Agent Development
Now available on GitHub, the SDK was initially created for internal use to speed up agent deployment, reducing timelines from months to days
Amazon Web Services (AWS) has open-sourced the Strands Agents SDK, an open-source toolkit designed to simplify the development of AI agents.
Now available on GitHub, the SDK was initially created for internal use to speed up agent deployment, reducing timelines from months to days.
"Multiple teams at AWS already use Strands for their AI agents in production, including Amazon Q Developer, AWS Glue, and VPC Reachability Analyzer," AWS said in a blog post.
The toolkit supports multiple large language models (LLMs) via Amazon Bedrock, Anthropic’s API, and open-source tools like Ollama and LiteLLM. This flexibility enables developers to integrate LLMs into third-party apps using thousands of available MCP clients.
Strands includes over 20 prebuilt tools, such as Retrieve—powered by Amazon Bedrock Knowledge Bases—and supports multi-agent collaboration.
A forthcoming update will introduce the Agent2Agent (A2A) protocol, allowing agents to share data and coordinate tasks.
The SDK also manages agent data and optimises storage to cut infrastructure costs.
"Strands Agents is an open community, and we’re excited that several companies are joining us with support and contributions, including Accenture, Anthropic, Langfuse, mem0.ai, Meta, PwC, Ragas.io, and Tavily," AWS added.
Earlier, Microsoft said it would embrace open AI agent collaboration with the upcoming integration of the Agent2Agent (A2A) protocol into its Azure AI Foundry and Copilot Studio.
The move aims to support seamless interoperability between AI agents across platforms, clouds, and organisations.
Comments ()