AWS Emulator: Free Local AWS Built by AI Agents

3D render of a small cloud server floating above a laptop on a desk, with the words "AWS ON YOUR LAPTOP."

Amazon Web Services has released a free tool that runs a copy of its cloud on your own laptop. That alone would be a useful story for developers. The real headline is who built it: autonomous AI agents wrote every line of code, and the human team says it neither wrote nor reviewed any of it.

AWS Emulator became generally available on 11 November 2026. It lets developers — and the AI coding agents working for them — test code against AWS services without deploying to the cloud, without a network connection, and without paying for cloud resources.

What AWS Emulator does

The emulator is a local stand-in for AWS. It is built from 438 AWS API models covering more than 200 services, and it serves requests on your own machine through a local HTTPS endpoint, by default at 127.0.0.1:9297.

The AWS CLI, the AWS SDKs, and coding agents such as Kiro and Claude Code connect to it through environment variables, so your application code stays unchanged. A test that creates a DynamoDB table, writes an item, and reads it back runs on the laptop, without a deployment.

There is no additional service fee. You still pay for AWS resources when you actually deploy to the cloud, but the local testing itself costs nothing. The code is public on GitHub under a CC0 public-domain dedication, with its full commit history visible.

How AI agents built it

The project started as a weekend experiment. An AWS engineer prompted the Kiro coding agent to build a passable version of Amazon SQS — the message queue service — in about 15 minutes. It worked well enough to grow into a real product.

Three teams of autonomous agents, running on Amazon Bedrock AgentCore, now build and maintain the emulator. The first team writes fidelity tests: short real-world workflows, such as creating a queue, sending a message, and deleting the queue, checked against real AWS. The second team runs those tests on the emulator and finds where its behaviour differs from the real cloud. The third team fixes the gaps.

The human team built the mechanisms that let the agents work on their own, and the quality-control gates around them — but, in the words of senior principal engineer Matthew Goldberg, "The emulator was written entirely by autonomous agents. Our team hasn't written or reviewed any of the code."

What it cannot do

AWS is unusually open about the limits, and they matter. The emulator does not evaluate IAM or resource-based policies the way real AWS does, so a request that AWS would deny may succeed locally. AWS says explicitly not to use it to test whether an application is secure.

Each emulator acts as a single account, with no cross-account or cross-region behaviour. It does not yet simulate throttling, outages, or differences in which services are available by region, and it does not sandbox the code it runs. An operation that is not implemented yet returns HTTP 501, "Not Implemented", so failures are loud rather than silent — which is the right design for a test tool.

Coverage is also uneven across services. AWS publishes a fidelity score per service, and the public repository has a folder literally called "things-that-do-not-work" listing known gaps. Check that folder before trusting a local test result.

Why this matters for coding agents

Coding agents made this tool necessary. Agents such as Kiro and Claude Code can write code far faster than teams can test it against the real cloud. Every test used to mean a deployment, a wait, and a bill.

The emulator closes that loop on the developer's own machine. An agent can create a resource, inspect the result, and revise its code without waiting for a cloud deployment.

It is a genuine milestone: a production developer tool from the world's largest cloud provider, maintained by agents, with its gaps published in public.

Key takeaways

  • AWS Emulator launched on 11 November 2026: a free local emulator for more than 200 AWS services, built from 438 API models.
  • Developers and coding agents test against it on their own machine, with no AWS account, network, or extra fee needed.
  • Autonomous AI agents wrote every line of code; the human team neither wrote nor reviewed any of it.
  • Three agent teams maintain it: fidelity-test writers, gap finders, and fixers, with automated quality gates.
  • It does not test IAM policies, has no cross-region behaviour, and does not simulate throttling or outages. Known gaps are listed publicly.

Conclusion

AWS Emulator solves a real developer problem — testing cloud code without the cloud — and it does it with a story that would have sounded like science fiction two years ago: a major cloud tool written entirely by AI agents, maintained by AI agents, with its flaws published for everyone to see. Try it for local testing and agent workflows. Just remember what it is not: it is not a copy of production, and it is not a security test. Check the "things-that-do-not-work" folder, test IAM on real AWS, and enjoy the free local cloud.

FAQ

Is AWS Emulator really free?

Yes. There is no additional service fee, and it runs on your own machine without an AWS account. You only pay for AWS resources when you deploy to the real cloud.

Which services does AWS Emulator cover?

It is built from 438 AWS API models covering more than 200 services. Coverage is uneven — AWS publishes a fidelity score per service, and unimplemented operations return HTTP 501.

Can AWS Emulator test my app's security?

No. It does not evaluate IAM or resource-based policies the way real AWS does, so a request AWS would deny may succeed locally. AWS says not to use it for security testing.

Researched with AI assistance and reviewed for accuracy.

Sources

Follow on WhatsApp

Follow Daily Tech & AI on WhatsApp for the latest tech and AI news, every day: Daily Tech & AI on WhatsApp

Comments