This is a proof of concept to see if we can replicate Amazon API Gateway using docker images to run lambda
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Serverless simulation plugin

serverless Build Status npm version Coverage Status dependencies license

This is a proof of concept to see if we can replicate Amazon API Gateway using docker images to run lambda


  • λ runtimes supported by docker-lambda.
  • CORS
  • Authorizer
    • Custom Authorizer supported
    • Coginito Authorizer not implemented yet
  • Lambda Integration
    • Velocity templates support. supported
  • Lambda Proxy Integration. supported


Getting Started

Install the plugin

npm i --save-dev serverless-plugin-simulate

Configure your service to use the plugin

service: my-service
  name: aws
  runtime: nodejs4.3 # python2.7 is also supported

# this is optional configurations
# servicesPathDest is use for the case u want to compile all your src and out
# put to another folder like dist
# services allows specifying a docker-compose.yml file and (optional)
# projectName. This will start the docker-compose stack when simulate is run.
# If projectName is specified, it will be used when running docker-compose,
# and the default docker-compose network will be passed to lambda docker commands
# allowing all lambdas to access any hosts defined in docker-compose services
    dist: dist
      file: docker-compose.yml
      # will use: $ docker-compose --project-name myproject ...
      # and:      $ docker --network myproject_default ...
      projectName: myproject

  - serverless-plugin-simulate

If you do not need to chain functions locally you can just run the API Gateway simulation by itself.

sls simulate apigateway -p 5000

Using the Lambda simulator

If you want to chain functions locally, you need to use the Lambda Simulator.

Run the Lambda Simulation

sls simulate lambda -p 4000

Run the API Gateway Simulation

sls simulate apigateway -p 5000 --lambda-port 4000

Use the environment variables to configure the AWS SDK to use the local Lambda simulation. You can use the same technique with any other AWS SDK.

const AWS = require('aws-sdk');

const endpoint = process.env.SERVERLESS_SIMULATE ?

const lambda = new AWS.Lambda({ endpoint })
const handler = (event, context, callback) => {
  const params = {
    FunctionName: 'my-other-function',
    Payload: JSON.stringify({ foo: 'bar' })
  lambda.invoke(params, (err, result) => {
    if (err) {
      return callback(err)

    callback(null, {
      statusCode: 200,
      body: result.Payload


See the examples folder for examples.

  • npm install - Installs all dependencies
  • npm start - Starts API Gateway simulation listening at http://localhost:5000
  • npm run start:lambda - Starts Lambda simulation listening at http://localhost:4000
  • npm run start:apigateway - Starts API Gateway simulation that uses the Lambda simulation listening at http://localhost:5000
  • npm run start:services - Starts mock services defined in docker-compose.yml
  • npm test - tests custom authorizer (Authorization:TOKEN 12345)


See also the list of contributors who participated in this project.


Please create an issue before submitting an Pull Request.


This would not be possible without lambci

  • docker-lambda - Docker images and test runners that replicate the live AWS Lambda environment

@johncmckim for suggesting the idea