Give Your AI Agent a Local AWS Environment with LocalStack
Accelerate your agentic workflows with a local AWS sandbox
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds.
Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks.
In an agentic workflow, that means giving AI access to a public cloud account, racking up costs on the AWS bill, and waiting for provisioning to complete every time you push new code to the environment.
Join this livestreamed demo on Thursday, June 25, to see how our users:
- Give AI agents a simulated AWS environment that runs locally so they can test for the cloud without access or costs
- Enable agents to deploy and validate code in the local environment
- Review failures, read logs, and validate fixes instantly directly via their agents
- Ship validated code to staging or production by simply swapping the endpoint for your AWS account
Accelerate your agentic workflows with a local AWS sandbox
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds.
Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks.
In an agentic workflow, that means giving AI access to a public cloud account, racking up costs on the AWS bill, and waiting for provisioning to complete every time you push new code to the environment.
Join this livestreamed demo on Thursday, June 25, to see how our users:
- Give AI agents a simulated AWS environment that runs locally so they can test for the cloud without access or costs
- Enable agents to deploy and validate code in the local environment
- Review failures, read logs, and validate fixes instantly directly via their agents
- Ship validated code to staging or production by simply swapping the endpoint for your AWS account
Get Involved
Join Our Slack Community
Join our Meetup Group
Submit a Talk
Docs
Launch yourself in the world of local cloud development
Accelerate your agentic workflows with a local AWS sandbox
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds.
Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks.
In an agentic workflow, that means giving AI access to a public cloud account, racking up costs on the AWS bill, and waiting for provisioning to complete every time you push new code to the environment.
Join this livestreamed demo on Thursday, June 25, to see how our users:
- Give AI agents a simulated AWS environment that runs locally so they can test for the cloud without access or costs
- Enable agents to deploy and validate code in the local environment
- Review failures, read logs, and validate fixes instantly directly via their agents
- Ship validated code to staging or production by simply swapping the endpoint for your AWS account
Talks
Give Your AI Agent a Local AWS Environment with LocalStack
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds. Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks. That is slow, and on cloud services a wrong guess by an agent can get expensive.
LocalStack gives the agent somewhere to check its own work first. Same AWS APIs, running locally, no cloud account in the loop. The agent deploys, reads the logs, fixes what failed, and only suggests pushing to the cloud once it has seen the thing run.
Brian opens with how agent support works in LocalStack today, including the /ai entry point that lets an agent install and start LocalStack from a single prompt. He then covers Skills: six pre-built workflows an agent can run for container lifecycle, IaC deployment, state management, log analysis, IAM policy generation, and extensions. Each is a fixed set of steps the agent follows instead of improvising.
Harsh covers the MCP server and the tools that let an agent operate LocalStack directly rather than just printing commands for you to run. He walks through a testing loop where the agent deploys infrastructure, inspects traces and logs, injects faults to check resilience, and generates least-privilege IAM policies, all on the local environment. The aim is simple: get agents doing real verification before deploy, not afte
Give Your AI Agent a Local AWS Environment with LocalStack
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds. Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks. That is slow, and on cloud services a wrong guess by an agent can get expensive.
LocalStack gives the agent somewhere to check its own work first. Same AWS APIs, running locally, no cloud account in the loop. The agent deploys, reads the logs, fixes what failed, and only suggests pushing to the cloud once it has seen the thing run.
Brian opens with how agent support works in LocalStack today, including the /ai entry point that lets an agent install and start LocalStack from a single prompt. He then covers Skills: six pre-built workflows an agent can run for container lifecycle, IaC deployment, state management, log analysis, IAM policy generation, and extensions. Each is a fixed set of steps the agent follows instead of improvising.
Harsh covers the MCP server and the tools that let an agent operate LocalStack directly rather than just printing commands for you to run. He walks through a testing loop where the agent deploys infrastructure, inspects traces and logs, injects faults to check resilience, and generates least-privilege IAM policies, all on the local environment. The aim is simple: get agents doing real verification before deploy, not afte
Shelton Graves is a Senior Technical Product Manager at LocalStack, focused on Developer Experience. Before joining LocalStack, he served as Senior Product Manager for Docker Hub, and brings over a decade of experience across cloud native infrastructure, Kubernetes, serverless, and developer tooling.
Brian Rinaldi leads the Developer Relations team at LocalStack. Brian has over 25 years of experience as a developer – mostly for the web – and over a decade in Developer Relations for companies like Adobe, Progress Software, and LaunchDarkly.
Harsh is an Engineer at LocalStack and an AWS Community Builder. His interests are in DevOps, Platform Engineering, and CI/CD pipelines.
Get Involved
Join Our Slack Community
Join our Meetup Group
Submit a Talk
Docs
Launch yourself in the world of local cloud development
Give Your AI Agent a Local AWS Environment with LocalStack
Accelerate your agentic workflows with a local AWS sandbox
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds.
Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks.
In an agentic workflow, that means giving AI access to a public cloud account, racking up costs on the AWS bill, and waiting for provisioning to complete every time you push new code to the environment.
Join this livestreamed demo on Thursday, June 25, to see how our users:
- Give AI agents a simulated AWS environment that runs locally so they can test for the cloud without access or costs
- Enable agents to deploy and validate code in the local environment
- Review failures, read logs, and validate fixes instantly directly via their agents
- Ship validated code to staging or production by simply swapping the endpoint for your AWS account
Schedule
Give Your AI Agent a Local AWS Environment with LocalStack
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds. Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks. That is slow, and on cloud services a wrong guess by an agent can get expensive.
LocalStack gives the agent somewhere to check its own work first. Same AWS APIs, running locally, no cloud account in the loop. The agent deploys, reads the logs, fixes what failed, and only suggests pushing to the cloud once it has seen the thing run.
Brian opens with how agent support works in LocalStack today, including the /ai entry point that lets an agent install and start LocalStack from a single prompt. He then covers Skills: six pre-built workflows an agent can run for container lifecycle, IaC deployment, state management, log analysis, IAM policy generation, and extensions. Each is a fixed set of steps the agent follows instead of improvising.
Harsh covers the MCP server and the tools that let an agent operate LocalStack directly rather than just printing commands for you to run. He walks through a testing loop where the agent deploys infrastructure, inspects traces and logs, injects faults to check resilience, and generates least-privilege IAM policies, all on the local environment. The aim is simple: get agents doing real verification before deploy, not afte
Get Involved
Join Our Slack Community
Join our Meetup Group
Submit a Talk
Docs
Launch yourself in the world of local cloud development
