AWS API MCP Server v1.0.0 release (aws.amazon.com)

AWS released version 1.0.0 of the AWS API Model Context Protocol (MCP) Server, a milestone that lets foundation models interact with AWS services via natural language by generating and executing syntactically correct AWS CLI commands. The release focuses on production readiness and developer ergonomics: faster startup, fewer local dependencies by converting suggest_aws_command into a remote service, and new deployment options including streamable HTTP transport (in addition to stdio) and a container image in the Amazon ECR Public Gallery. The server is open-source on the AWS Labs GitHub and can be registered with popular MCP registries and integrated into agentic frameworks and MCP-compatible clients. For the AI/ML community this matters because it lowers friction for building LLM-driven agents and human-in-the-loop workflows that operate on cloud resources while adding operational controls. New security and stability features include improved secure file-system controls, stricter input validation, CloudWatch logging for observability, elicitation support for iterative prompts, and configurable safeguards to deny risky actions or require human consent for mutating operations. An experimental get_execution_plan tool (enabled via EXPERIMENTAL_AGENT_SCRIPTS) can produce prescriptive workflows for common AWS tasks, further enabling safer, more auditable automation driven by models.