AWS Emulator: AI Agents Build Open-Source Tool for Local Cloud Testing

Amazon Web Services (AWS) has introduced AWS Emulator, a new open-source tool designed to simulate AWS services locally on a developer's machine, within CI pipelines, or on development servers. A remarkable aspect of this release is that the entire tool was developed by autonomous AI agents, with no code written or reviewed by human AWS engineers. This initiative aims to provide developers with a cost-effective way to test their AWS applications without incurring cloud service fees during the development phase.

The development process involved three distinct teams of AI agents, each assigned specific roles to ensure the emulator's functionality and fidelity. The first team was responsible for generating tests based on publicly available AWS API specifications and validating these tests against actual AWS services. This comprehensive suite of tens of thousands of tests can be replayed without needing to contact AWS, ensuring consistent testing scenarios. The second team utilized these tests to identify discrepancies between the emulator's behavior and that of real AWS services, reporting and grouping failures that shared common underlying causes.

The third team of AI agents was tasked with fixing the identified issues, with each agent focusing on a single problem. Before any code changes could be integrated into the main codebase, they underwent rigorous checks by other agents, including assessments for coding standards, security vulnerabilities, and functional fidelity. Any changes that failed these checks were sent back for revision, ensuring a high level of code quality and adherence to AWS standards. AWS reports that hundreds of agents work on the code concurrently, addressing thousands of issues daily.

Developers can integrate AWS Emulator into their workflows in several ways, including via an AI coding agent, a Linux package, Docker, or an embedded mode. The Linux package operates as a local process, while Docker provides support across Windows, macOS, and Linux. Embedded mode allows the emulator to run directly within a Rust test process. This flexibility enables developers to test the interaction of their code with various AWS services, including cross-service calls, directly on their local machines.

Furthermore, the emulator facilitates running integration tests within CI pipelines without requiring an AWS account or dedicated cloud resources. It also supports prototyping applications locally with AI coding agents before final deployment on AWS. However, AWS Emulator is explicitly not intended for production workloads. It does not evaluate AWS Identity and Access Management (IAM) or resource-based policies, making it unsuitable for security testing.

Users are advised to use a single region and account for all requests to an emulator instance, as it does not segregate resources by region or account. The fidelity of the emulation varies by service and feature, with AWS checking behavior against its US West (Oregon) region, except for global services tested in US East (N. Virginia). All emulated resources are stored in memory and are lost when the emulator stops; therefore, it has not been tested for long-running use or load testing.

By default, the emulator collects anonymous usage and error data to send to AWS, though this reporting can be disabled. Crucially, it does not transmit any request, response, or resource data to AWS. The source code and fidelity tests are publicly available on GitHub under the CC0 public-domain dedication, allowing for community contributions and transparency. AWS also publishes per-service fidelity scores with each release, providing insight into the emulator's accuracy.