Browse by type
Scorekeep is a RESTful web API implemented in Java that uses Spring to provide an HTTP interface for creating and managing game sessions and users. This project includes the Scorekeep API and a front-end web app that consumes it. The front end and API can run on the same server and domain or separately, with the API running in Elastic Beanstalk and the front end served statically by a CDN.
The master branch shows the use of Spring, Angular, nginx, the AWS SDK for Java, Amazon DynamoDB, Gradle, CORS, and AWS Elastic Beanstalk features that enable you to:
Other branches extend the application's functionality and show the use of other AWS services. See the readme in each branch for details about the integration and instructions for use.
Branches
- cognito - Support login and store users in an Amazon Cognito user pool. Get AWS SDK credentials and make service calls with a Cognito identity pool.
- cognito-basic - Use Cognito for user ID storage. User pool only, no identity pool.
- lambda - Call an AWS Lambda function to generate random names.
- lambda-worker - Run a Lambda function periodically to process game records and store the output in Amazon S3.
- sql - Use JDBC to store game histories in an attached PostgreSQL database instance.
- xray - Use the AWS X-Ray SDK for Java to instrument incoming requests, functions, SDK clients, SQL queries, HTTP clients, startup code, and AWS Lambda functions.
- xray-cognito - Use AWS credentials obtained with Amazon Cognito to upload trace data to X-Ray from the browser.
- xray-gettingstarted (tutorial) - Use the AWS X-Ray SDK for Java to instrument incoming requests and SDK clients (no additional configuration required).
- xray-worker - Instrumented Python Lambda worker function from the lambda-worker branch.
Use the procedures in the following sections to run the project on Elastic Beanstalk and configure it for local testing and development.
Sections - Prerequisites - Deploying the application - Configuring notifications - How it works - Running the project locally - Contributing
If you're using an IAM user with limited permissions, good work! Add Elastic Beanstalk permissions to your user account to get started.
To add Elastic Beanstalk permissions to an IAM user
Create a Java 8 SE environment in Elastic Beanstalk to host the application.
To create an Elastic Beanstalk environment running the Java 8 SE platform
When the Scorekeep API runs in Elastic Beanstalk, it uses the permissions of its EC2 instance to call AWS. Elastic Beanstalk provides a default instance profile that you can extend to grant the application the permissions it needs to read from and write to resource tables in DynamoDB, and send notifications with SNS.
To add DynamoDB and SNS permissions to the instances in your Elastic Beanstalk environment
Deploy the source code for the project to your Elastic Beanstalk environment.
To deploy the source code

Click through the app to explore its functionality. Use the network console in your browser to see the HTTP requests that it sends to the API to read and write users, sessions, games, moves, and game state to DynamoDB via the API.
The API uses SNS to send a notification email when a game ends. To enable notifications, configure your email address in an environment variable.
To enable notifications
The project includes two independent components
The API runs at paths under /api that provide access to user, session, game, state, and move resources stored as JSON documents in DynamoDB. The API is RESTful, so you can create resources by sending HTTP POST requests to the resource path, for example /api/session. See the test script for example requests with cURL.
The application includes a configuration file that creates a DynamoDB table for each resource type.
The Buildfile tells Elastic Beanstalk to run gradle build during deployment to create an executable JAR file. The Procfile tells Elastic Beanstalk to run that JAR on port 5000.
The front end is an Angular 1.5 web app that uses $resource objects to perform CRUD operations on resources defined by the API. Users first encounter the main view and controller and progress through session and game views at routes that include the IDs of resources that the user creates.
The front end is served statically by the nginx proxy at the root path. The nginx.conf file in the source code overwrites the default configuration provided by the Elastic Beanstalk Java platform to serve the front end statically, and forward requests to paths starting with /api to the API running on port 5000.
You can run both the API and front end locally with Gradle and the Spring Boot CLI. To get started, clone this repository or extract the contents of the source bundle that you downloaded earlier.
The API requires DynamoDB tables to exist in AWS to run locally. These tables are created by this configuration file when you launch the application in Elastic Beanstalk. If you terminate the environment, the tables are deleted. To create the tables without a running environment, use the configuration file as a template to create an AWS CloudFormation stack.
The application reads an environment variable named AWS_REGION to determine which region to connect to for calls to DynamoDB. In Elastic Beanstalk, this variable is set in a configuration file that reads the current region from CloudFormation and creates an Elastic Beanstalk environment property.
Set the environment variable and then use Gradle to build the API and run it locally on port 5000.
~/eb-java-scorekeep$ export AWS_REGION=us-west-2
~/eb-java-scorekeep$ gradle bootrun
The application needs AWS credentials to communicate with DynamoDB. In Elastic Beanstalk, Scorekeep gets credentials from the instance profile, which is the IAM role that is attached to the EC2 instance that runs the code. When you run the application locally, the AWS SDK for Java can retrieve credentials from files in ~/.aws/ or environment variables.
Follow the instructions in the AWS SDK for Java Developer Guide to provide access keys to the application: Set up AWS Credentials for Development.
Use the test script to verify that the API works.
~/eb-java-scorekeep$ ./bin/test-api.sh
The script targets localhost:5000. However, you can point it at the API running on any host by modifying the API variable at the top of the file.
Use the Spring Boot CLI to run the front end on port 8080.
~/eb-java-scorekeep$ spring run app.groovy
Open the app in a browser: localhost:8080
The app loads but can't hit the API, because it's trying to call paths relative to its own root, localhost:8080, but the API is running on localhost:5000 (or in Elastic Beanstalk). To fix this, configure the app with the full URL of the API.
To configure the web app with an absolute path to the API
Use the domain name of your environment to test changes to the front end without running the backend locally.
module.value('api', 'http://scorekeep.XXXXXXXX.elasticbeanstalk.com/api/');
Use localhost:5000 to test both front end and backend changes when running both locally.
module.value('api', 'http://localhost:5000/api/');
Refresh the app in your browser to load the updated script.
The API includes a CORS (cross-origin resource sharing) filter that allows traffic from localhost:8080.
private static UrlBasedCorsConfigurationSource configurationSource() {
CorsConfiguration config = new CorsConfiguration();
config.setAllowCredentials(true);
// Modify allowed origins if you run the client at a different endpoint
config.addAllowedOrigin("http://localhost:8080");
config.addAllowedHeader("*");
config.addAllowedMethod("*");
UrlBasedCorsConfigurationSource source = new UrlBasedCorsConfigurationSource();
source.registerCorsConfiguration("/**", config);
return source;
}
This lets requests originating from a front end hosted locally on port 8080 to send requests to the API hosted on a different local port (i.e., localhost:5000) or on Elastic Beanstalk. This enables you to work on the front end locally and see changes immediately, without needing to deploy the source to your environment. When you make changes to the HTML or CSS in the app, simply refresh the browser.
When you run both the
$ claude mcp add eb-java-scorekeep \
-- python -m otcore.mcp_server <graph>