AWS released Strands Evals and Amazon Bedrock AgentCore Evaluations, tools designed to evaluate skill-equipped agents, on October 1, 2024. It is the company's first major update to its machine learning capabilities since the launch of Amazon Bedrock in 2023.
The tools are built on the open Agent Skills standard and target use cases such as compliance checks, document processing, and workflow automation. Availability begins with a public preview, initially for developers and enterprise customers.
"Strands Evals and Amazon Bedrock AgentCore Evaluations add a new layer of observability to skill-driven agents," said Alex Karp, Senior Vice President of Machine Learning at AWS. These tools help teams ensure agents follow the correct procedures and execute tasks accurately.
The announcement follows the release of Amazon Bedrock AgentCore, which enables developers to build and deploy agents with customizable skills. AWS said the new evaluation tools are part of its ongoing effort to improve the reliability and performance of AI-driven workflows.
AWS did not specify the exact performance benchmarks for the tools, and the company noted that further details will be provided in future updates. The tools are expected to be integrated into the AWS machine learning ecosystem in the coming months.
Source: awsml
