This package has been deprecated

Author message:

AWS SAM CLI is not distributed via NPM anymore. Refer to the documentation for latest installation instructions -


0.2.11 • Public • Published

SAM Local

SAM Local (Beta)

Build Status Apache-2.0 Contributers GitHub-release npm-release Join the chat at

sam is the AWS CLI tool for managing Serverless applications written with AWS Serverless Application Model (SAM). SAM Local can be used to test functions locally, start a local API Gateway from a SAM template, validate a SAM template, and generate sample payloads for various event sources.

Main features

  • Develop and test your Lambda functions locally with sam local and Docker
  • Invoke functions from known event sources such as Amazon S3, Amazon DynamoDB, Amazon Kinesis, etc.
  • Start local API Gateway from a SAM template, and quickly iterate over your functions with hot-reloading
  • Validate SAM templates



Running Serverless projects and functions locally with SAM Local requires Docker to be installed and running. SAM Local will use the DOCKER_HOST environment variable to contact the docker daemon.

For macOS and Windows users: SAM local requires that the project directory (or any parent directory) is listed in Docker file sharing options.

Verify that docker is working, and that you can run docker commands from the CLI (e.g. ‘docker ps’). You do not need to install/fetch/pull any containers – SAM Local will do it automatically as required.

Windows, Linux, macOS with NPM [Recommended]

The easiest way to install sam is to use NPM.

npm install -g aws-sam-local

Verify the installation worked:

sam --version

If you get a permission error when using npm (such as EACCES: permission denied), please see the instructions on this page of the NPM documentation:

Upgrading via npm

To update sam once installed via npm:

npm update -g aws-sam-local

Binary release

We also release the CLI as binaries that you can download and instantly use. You can find them under Releases in this repo. In case you cannot find the version or architecture you're looking for you can refer to Build From Source section for build details.

Build From Source

First, install Go (v1.8+) on your machine:, then run the following:

$ go get 

This will install sam to your $GOPATH/bin folder. Make sure this directory is in your $PATH (or %%PATH%% on Windows) and you should then be able to use the SAM Local. Please note that due to the package name, the binary will be installed as aws-sam-local rather than sam.

aws-sam-local --help


sam requires a SAM template in order to know how to invoke your function locally, and it's also true for spawning API Gateway locally - If no template is specified template.yaml will be used instead.

You can find sample SAM templates either under samples located in this repo or by visiting SAM official repository.

Invoke functions locally

SAM Local Invoke Sample

You can invoke your function locally by passing its SAM logical ID and an event file. Alternatively, sam local invoke accepts stdin as an event too.

  Ratings:  # <-- Logical ID 
    Type: 'AWS::Serverless::Function'


# Invoking function with event file 
$ sam local invoke "Ratings" -e event.json
# Invoking function with event via stdin 
echo '{"message": "Hey, are you there?" }' | sam local invoke "Ratings"
# For more options 
$ sam local invoke --help

Generate sample event source payloads

To make local development and testing of Lambda functions easier, you can generate mock/sample event payloads for the following services:

  • S3
  • Kinesis
  • DynamoDB
  • Cloudwatch Scheduled Event
  • Cloudtrail
  • API Gateway


sam local generate-event <service>

Also, you can invoke an individual lambda function locally from a sample event payload - Here's an example using S3:

sam local generate-event s3 --bucket <bucket> --key <key> | sam local invoke <function logical id>

For more options, see sam local generate-event --help.

Run API Gateway locally

sam local start-api spawns a local API Gateway to test HTTP request/response functionality. Features hot-reloading to allow you to quickly develop, and iterate over your functions.

SAM Local Start API


sam local start-api

sam will automatically find any functions within your SAM template that have Api event sources defined, and mount them at the defined HTTP paths.

In the example below, the Ratings function would mount at /ratings for GET requests.

  Type: AWS::Serverless::Function
    Handler: ratings.handler
    Runtime: python3.6
        Type: Api
          Path: /ratings
          Method: get

By default, SAM uses Proxy Integration and expects the response from your Lambda function to include one or more of the following: statusCode, headers and/or body.

For example:

// Example of a Proxy Integration response
exports.handler = (event, context, callback) => {
    callback(null, {
        statusCode: 200,
        headers: { "x-custom-header" : "my custom header value" },
        body: "hello world"

For examples in other AWS Lambda languages, see this page.

If your function does not return a valid Proxy Integration response then you will get a HTTP 500 (Internal Server Error) when accessing your function. SAM Local will also print the following error log message to help you diagnose the problem:

ERROR: Function ExampleFunction returned an invalid response (must include one of: body, headers or statusCode in the response object)

Debugging Applications

Both sam local invoke and sam local start-api support local debugging of your functions.

To run SAM Local with debugging support enabled, just specify --debug-port or -d on the command line.

# Invoke a function locally in debug mode on port 5858  
$ sam local invoke -d 5858 <function logical id> 
# Start local API Gateway in debug mode on port 5858 
$ sam local start-api -d 5858

Note: If using sam local start-api, the local API Gateway will expose all of your Lambda functions but, since you can specify a single debug port, you can only debug one function at a time. You will need to hit your API before SAM Local binds to the port allowing the debugger to connect.

Here is an example showing how to debug a NodeJS function with Microsoft Visual Studio Code:

SAM Local debugging example

In order to setup Visual Studio Code for debugging with AWS SAM Local, use the following launch configuration:

    "version": "0.2.0",
    "configurations": [
            "name": "Attach to SAM Local",
            "type": "node",
            "request": "attach",
            "address": "localhost",
            "port": 5858,
            "localRoot": "${workspaceRoot}",
            "remoteRoot": "/var/task",
            "protocol": "legacy"

Note: Node.js versions below 7 (e.g. Node.js 4.3 and Node.js 6.10) use the legacy protocol, while Node.js versions including and above 7 (e.g. Node.js 8.10) use the inspector protocol. Be sure to specify the corresponding protocol in the protocol entry of your launch configuration.

Debugging Python functions

Unlike Node.JS and Java, Python requires you to enable remote debugging in your Lambda function code. If you enable debugging with --debug-port or -d for a function that uses one of the Python runtimes, SAM Local will just map through that port from your host machine through to the Lambda runtime container. You will need to enable remote debugging in your function code. To do this, use a python package such as remote-pdb. When configuring the host the debugger listens on in your code, make sure to use not to allow Docker to map through the port to your host machine.

Please note, due to a open bug with Visual Studio Code, you may get a Debug adapter process has terminated unexpectedly error when attempting to debug Python applications with this IDE. Please track the GitHub issue for updates.

Passing Additional Runtime Debug Arguments

To pass additional runtime arguments when debugging your function, use the environment variable DEBUGGER_ARGUMENTS. This will pass a string of arguments directly into the run command SAM Local uses to start your function.

For example, if you want to load a debugger like iKPdb at runtime of your Python function, you could pass the following as DEBUGGER_ARGUMENTS: -m ikpdb --ikpdb-port=5858 --ikpdb-working-directory=/var/task/ --ikpdb-client-working-directory=/myApp --ikpdb-address= This would load iKPdb at runtime with the other arguments you've specified. In this case, your full SAM local command would be:

$ DEBUGGER_ARGUMENTS="-m ikpdb --ikpdb-port=5858 --ikpdb-working-directory=/var/task/ --ikpdb-client-working-directory=/myApp --ikpdb-address=" echo {} | sam local invoke -d 5858 myFunction

You may pass debugger arguments to functions of all runtimes.

Connecting to docker network

Both sam local invoke and sam local start-api support connecting the create lambda docker containers to an existing docker network.

To connect the containers to an existing docker network, you can use the --docker-network command-line argument or the SAM_DOCKER_NETWORK environment variable along with the name or id of the docker network you wish to connect to.

# Invoke a function locally and connect to a docker network 
$ sam local invoke --docker-network my-custom-network <function logical id>
# Start local API Gateway and connect all containers to a docker network 
$ sam local start-api --docker-network b91847306671 -d 5858

Validate SAM templates

Validate your templates with $ sam validate. Currently this command will validate that the template provided is valid JSON / YAML. As with most SAM Local commands, it will look for a template.yaml file in your current working directory by default. You can specify a different template file/location with the -t or --template option.


$ sam validate

Note: More in-depth functionality is currently disabled. An alternative validation route is to validate your JSON against schema for the whole CloudFormation and SAM specification.

Package and Deploy to Lambda

Once you have developed and tested your Serverless application locally, you can deploy to Lambda using sam package and sam deploy command. package command will zip your code artifacts, upload to S3 and produce a SAM file that is ready to be deployed to Lambda using AWS CloudFormation. deploy command will deploy the packaged SAM template to CloudFormation. Both sam package and sam deploy are identical to their AWS CLI equivalents commands aws cloudformation package and aws cloudformation deploy respectively. Please consult the AWS CLI command documentation for usage.


# Package SAM template 
$ sam package --template-file sam.yaml --s3-bucket mybucket --output-template-file packaged.yaml
# Deploy packaged SAM template 
$ sam deploy --template-file ./packaged.yaml --stack-name mystack --capabilities CAPABILITY_IAM

Getting started


Compiled Languages (Java)

To use SAM Local with compiled languages, such as Java that require a packaged artifact (e.g. a JAR, or ZIP), you can specify the location of the artifact with the AWS::Serverless::Function CodeUri property in your SAM template.

For example:

AWSTemplateFormatVersion: 2010-09-09
Transform: AWS::Serverless-2016-10-31

    Type: AWS::Serverless::Function
      Handler: com.example.HelloWorldHandler
      CodeUri: ./target/HelloWorld-1.0.jar
      Runtime: java8

You should then build your JAR file using your normal build process. Please note that JAR files used with AWS Lambda should be a shaded JAR file (or uber jar) containing all of the function dependencies.

// Build the JAR file
$ mvn package shade:shade

// Invoke with SAM Local
$ echo '{ "some": "input" }' | sam local invoke

// Or start local API Gateway simulator
$ sam local start-api

You can find a full Java example in the samples/java folder

IAM Credentials

SAM Local will invoke functions with your locally configured IAM credentials.

As with the AWS CLI and SDKs, SAM Local will look for credentials in the following order:

  1. Environment Variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY).
  2. The AWS credentials file (located at ~/.aws/credentials on Linux, macOS, or Unix, or at C:\Users\USERNAME \.aws\credentials on Windows).
  3. Instance profile credentials (if running on Amazon EC2 with an assigned instance role).

In order to test API Gateway with a non-default profile from your AWS credentials file append --profile <profile name> to the start-api command:

// Test API Gateway locally with a credential profile.
$ sam local start-api --profile some_profile

See this Configuring the AWS CLI for more details.

Lambda Environment Variables

If your Lambda function uses environment variables, you can provide values for them will passed to the Docker container. Here is how you would do it:

For example, consider the SAM template snippet:

    Type: AWS::Serverless::Function
      Handler: index.handler
      Runtime: nodejs4.3
          TABLE_NAME: prodtable
          BUCKET_NAME: prodbucket
    Type: AWS::Serverless::Function
      Handler: app.handler
      Runtime: nodejs4.3
          STAGE: prod
          TABLE_NAME: prodtable

Environment Variable file

Use --env-vars argument of invoke or start-api commands to provide a JSON file that contains values for environment variables defined in your function. The file should be structured as follows:

  "MyFunction1": {
    "TABLE_NAME": "localtable",
    "BUCKET_NAME": "testBucket"
  "MyFunction2": {
    "TABLE_NAME": "localtable",
    "STAGE": "dev"
$ sam local start-api --env-vars env.json

in alternative you can pass a cloudformation configuration.json containing a parameters key:

  "Parameters": {
    "TABLE_NAME": "localtable",
    "BUCKET_NAME": "testBucket"

Shell environment

Variables defined in your Shell's environment will be passed to the Docker container, if they map to a Variable in your Lambda function. Shell variables are globally applicable to functions ie. If two functions have a variable called TABLE_NAME, then the value for TABLE_NAME provided through Shell's environment will be availabe to both functions.

Following command will make value of mytable available to both MyFunction1 and MyFunction2

$ TABLE_NAME=mytable sam local start-api

Combination of Shell and Environment Variable file

For greater control, you can use a combination shell variables and external environment variable file. If a variable is defined in both places, the one from the file will override the shell. Here is the order of priority, highest to lowest. Higher priority ones will override the lower.

  1. Environment Variable file
  2. Shell's environment
  3. Hard-coded values from the template

Identifying local execution from Lambda function code

When your Lambda function is invoked using SAM Local, it sets an environment variable AWS_SAM_LOCAL=true in the Docker container. Your Lambda function can use this property to enable or disable functionality that would not make sense in local development. For example: Disable emitting metrics to CloudWatch (or) Enable verbose logging etc.

Static Assets

Often, it's useful to serve up static assets (e.g CSS/HTML/Javascript etc) when developing a Serverless application. On AWS, this would normally be done with CloudFront/S3. SAM Local by default looks for a ./public/ directory in your SAM project directory and will serve up all files from it at the root of the HTTP server when using sam local start-api. You can override the default static asset directory by using the -s or --static-dir command line flag. You can also disable this behaviour completely by setting --static-dir "".

Local Logging

Both invoke and start-api command allow you to pipe logs from the function's invocation into a file. This will be useful if you are running automated tests against SAM Local and want to capture logs for analysis.


$ sam local invoke --log-file ./output.log

Remote Docker

Sam Local loads function code by mounting filesystem to a Docker Volume. As a result, The project directory must be pre-mounted on the remote host where the Docker is running.

If mounted, you can use the remote docker normally using --docker-volume-basedir or environment variable SAM_DOCKER_VOLUME_BASEDIR.

Example - Docker Toolbox (Windows):

When you install and run Docker Toolbox, the Linux VM with Docker is automatically installed in the virtual box.

The /c/ path for this Linux VM is automatically shared with C:\ on the host machine.

sam local invoke --docker-volume-basedir /c/Users/shlee322/projects/test "Ratings"

Project Status

  • Supported AWS Lambda Runtimes
    • nodejs
    • nodejs4.3
    • nodejs6.10
    • nodejs8.10
    • java8
    • python2.7
    • python3.6
    • dotnetcore1.0
    • dotnetcore2.0
  • AWS credential support
  • Debugging support
  • Inline Swagger support within SAM templates


Contributions and feedback are welcome! Proposals and pull requests will be considered and responded to. For more information, see the CONTRIBUTING file.

A special thank you

SAM Local uses the open source docker-lambda Docker images created by @mhart.


You can find sample functions code and a SAM template used in this README under the samples folder within this repo.




Package Sidebar


npm i aws-sam-local

Weekly Downloads






Unpacked Size

24.6 kB

Total Files


Last publish


  • amzn-oss
  • aws_sam
  • paulmaddox
  • sanathkr