Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


What is it?

Get live in minutes with a complete stack: react front end, lambdas, infrastructure and CICD pipeline.


Everyone wants to focus on the idea and not setup / config fiddling. A react front end with lambda backend is a common pattern for building web applications / startups. However, there is hours/days of setup to create a complete and robust setup.

Awsboilerplate attempts to provide a complete and integrated boilerplate solution:

  1. Front end react/redux stack: leveraging the react-boilerplate project built via codebuild
  2. Lambda back end: including exemplars for common lambda patterns - hello world, background jobs, data pipelines
  3. Infrastructure: Domain registration (via route53), DNS, Naked domain redirect, Cloudfront, API Gateways
  4. Monitoring: via Cloudwatch
  5. Everything is Code: All elements are defined in code/CDK (application, infrastructure and pipeline)
  6. CICD pipeline: self-mutating pipeline which deploys everything
  7. Supports local development workflow (SAM is configured)
  8. Integration: All the components are wired together (build, config, infrastructure)

Live Demo includes the react front end calling a 'hello world' lambda. This application is deployed via the CICD pipeline.


  1. Existing AWS Account
  2. AWS CLI
  3. Python / Pip / Virtualenv


Run Locally

AWSBoilerplate can be run locally.

  1. Fork and checkout the project
  2. cd awsboilerplate/app/lambda_hello_world; sam local start-api
  3. cd awsboilerplate/www; npm start
  4. Open http://localhost:3000/

Run on AWS

  1. Fork THEN clone this project (you need to fork as AWS will pull the code from your github project)
  2. Create a python venv
virtualenv -p python3.8 .venv
pip install -r requirements.txt
  1. Install AWS CDK. Run cdk synth to test the tools are installed correctly
  2. Run cdk bootstrap --trust <AWS ACCOUNT ID> --cloudformation-execution-policies arn:aws:iam::aws:policy/AdministratorAccess (this command sets up cdk on your account)
  3. Log into github and create an access token to allow codepipeline to pull from the repository
  4. Add this access token to AWS secrets manager using the following command
aws secretsmanager create-secret --name awsboilerplate-github-oath-token \
    --description "awsboilerplate Git Auth Token" \
    --region us-east-1
    --secret-string <YOUR OATH TOKEN>
  1. Register a domain via Route53 which will create a hosted zone. Run aws route53 list-hosted-zones. Take the hosted zone ID and domain name and enter them in
hosted_zone_name = "<your domain name i.e.>"
website_domain_name = "<Your Hosted Zone ID>"
  1. Update the server URL in awsboilerplate/www/internals/webpack/
    new webpack.EnvironmentPlugin({
      SERVER_URL: 'https://api.<Your domain name>' // <-- Keep https://api prefix. An API Gateway will be created for this URL
  1. Commit and push (to your github fork)
  2. Run cdk deploy. CDK will create the pipeline defined in awsboilerplate/pipeline/ When this pipeline runs it will create all of the infrastructure defined in awsboilerplate/infra and app.

Development Cycle / Making Changes

You can develop the react front end or lambdas locally then deploy.

When ready to push simply run

-- usual git add, git commit --
git push origin master

The CICD will pick up the changes to both the application code and infrastructure and mutate both. Once the pipeline runs you can hit the domain name to see the new changes (remember cloudfront cache may be enabled).

Architecture Overview

The project implements the serverless web application pattern.

  1. React front end (leveraging and integrating the react boilerplate)
  2. Back end lambdas
  3. CICD pipeline to deploy everything (mixing codepipeline, codebuild and cdkpipelines)


'Everything is Code'

awsboilerplate has some novelty. The project leverages CDK and a recently released feature: CDKPipelines.

awsboilerplate includes a custom pipeline build ontop of these technologies.

  1. Every element (front end, back end, cicd, infrastructure, monitoring) is implemented in code, in a single mono repository.
  2. It is possible to commit a change to any element and the pipeline should auto-magically make it happen (including changes to the pipeline itself)

We are calling this pattern 'Everything is Code' which is further detailed here


  1. 'Everything is Code'
  2. Alternatives
  3. Roadmap
  4. Limitations


awsboilerplate assemble and integrates multiple technologies into a full architecture/toolkit. The only novelty is stretching CDK to support a single mono-repository and combined pipeline.


  1. react-boilerplate and @mxstbr. awsboilerplate obviously incorporates the react-boilerplate as a component. The react boilerplate provided additional inspiration in the ability to solve a ton of developer friction and time (if you have ever tried to assemble a react-redux-webpack project from scratch)
  2. @rix0rrr provided invaluable help to answer a number of CDK questions
  3. @muymoo for answering every single other question!

Other Reading

  1. I found this too late into building awsboilerplate CDK Patterns
  2. Which is different from the official AWS pattern library


All contributions greatly appreciated. Trying to follow the edict from Reid Hoffman.

If you are not embarrassed by the first version of your product, you’ve launched too late

The launch version of the project is 'good enough' and solves many hours of development time for people starting new project.

There are many many features which could be added. Please submit a pull request. Also see the features page for open features/issues.


Released under the MIT license. Copyright 2021 geod