How to add Cognito authentication to a serverless API
In this example we will look at how to add Cognito User Pool authentication to a serverless API using Serverless Stack (SST).
Requirements
- Node.js >= 10.15.1
- We’ll be using Node.js (or ES) in this example but you can also use TypeScript
- An AWS account with the AWS CLI configured locally
Create an SST app
Let’s start by creating an SST app.
$ npx create-serverless-stack@latest api-auth-cognito
$ cd api-auth-cognito
By default our app will be deployed to an environment (or stage) called dev
and the us-east-1
AWS region. This can be changed in the sst.json
in your project root.
{
"name": "api-auth-cognito",
"stage": "dev",
"region": "us-east-1"
}
Project layout
An SST app is made up of two parts.
-
stacks/
— App InfrastructureThe code that describes the infrastructure of your serverless app is placed in the
stacks/
directory of your project. SST uses AWS CDK, to create the infrastructure. -
src/
— App CodeThe code that’s run when your API is invoked is placed in the
src/
directory of your project.
Setting up the API
Let’s start by setting up an API.
Replace the stacks/MyStack.js
with the following.
import * as sst from "@serverless-stack/resources";
export default class MyStack extends sst.Stack {
constructor(scope, id, props) {
super(scope, id, props);
// Create Api
const api = new sst.Api(this, "Api", {
defaultAuthorizationType: sst.ApiAuthorizationType.AWS_IAM,
routes: {
"GET /private": "src/private.main",
"GET /public": {
function: "src/public.main",
authorizationType: sst.ApiAuthorizationType.NONE,
},
},
});
// Show the API endpoint and other info in the output
this.addOutputs({
ApiEndpoint: api.url,
});
}
}
We are creating an API here using the sst.Api
construct. And we are adding two routes to it.
GET /private
GET /public
By default, all routes have the authorization type AWS_IAM
. This means the caller of the API needs to have the required IAM permissions. The first is a private endpoint. The second is a public endpoint and its authorization type is overriden to NONE
.
Setting up authentication
Add this below the sst.Api
definition in stacks/MyStack.js
.
// Create auth provider
const auth = new sst.Auth(this, "Auth", {
// Create a Cognito User Pool to manage user's authentication info.
cognito: {
userPool: {
// Users will login using their email and password
signInAliases: { email: true },
},
},
});
// Allow authenticated users to invoke the API
auth.attachPermissionsForAuthUsers([api]);
This creates a Cognito User Pool; a user directory that manages user sign up and login. We’ve configured the User Pool to allow users to login with their email and password.
This also creates a Cognito Identity Pool which assigns IAM permissions to users. We are allowing only the logged in users to have the permission to call the API.
Replace the this.addOutputs
call with the following.
this.addOutputs({
ApiEndpoint: api.url,
UserPoolId: auth.cognitoUserPool.userPoolId,
IdentityPoolId: auth.cognitoCfnIdentityPool.ref,
UserPoolClientId: auth.cognitoUserPoolClient.userPoolClientId,
});
We are going to print out the resources that we created for reference.
Adding function code
We will create two functions, one for the public route, and one for the private route.
Add a src/public.js
.
export async function main() {
return {
statusCode: 200,
body: "Hello stranger!",
};
}
Add a src/private.js
.
export async function main() {
return {
statusCode: 200,
body: "Hello user!",
};
}
Now let’s test our new API.
Starting your dev environment
SST features a Live Lambda Development environment that allows you to work on your serverless apps live.
$ npx sst start
The first time you run this command it’ll take a couple of minutes to do the following:
- It’ll bootstrap your AWS environment to use CDK.
- Deploy a debug stack to power the Live Lambda Development environment.
- Deploy your app, but replace the functions in the
src/
directory with ones that connect to your local client. - Start up a local client.
Once complete, you should see something like this.
===============
Deploying app
===============
Preparing your SST app
Transpiling source
Linting source
Deploying stacks
dev-api-auth-cognito-my-stack: deploying...
✅ dev-api-auth-cognito-my-stack
Stack dev-api-auth-cognito-my-stack
Status: deployed
Outputs:
UserPoolClientId: 4fb69je3470cat29p0nfm3t27k
UserPoolId: us-east-1_e8u3sktE1
ApiEndpoint: https://12mflx0e8e.execute-api.us-east-1.amazonaws.com
IdentityPoolId: us-east-1:d01df859-f416-4dc2-90ac-0c6fc272d197
The ApiEndpoint
is the API we just created. Make a note of the UserPoolClientId
, UserPoolId
, IdentityPoolId
; we’ll need them later.
Now let’s try out our public route. Head over to the following in your browser. Make sure to replace the URL with your API.
https://12mflx0e8e.execute-api.us-east-1.amazonaws.com/public
You should see the greeting Hello stranger!
.
And if you try to visit the private route, you will see {"message":"Forbidden"}
.
https://12mflx0e8e.execute-api.us-east-1.amazonaws.com/private
Signing up
Now to visit the private route, we need to create an account in our User Pool. Usually, we’ll have our users sign up for an account through our app. But for this example, we’ll use the AWS CLI to sign up a user and confirm their account.
Use the following command in your terminal. Replace --client-id
with UserPoolClientId
from the sst start
output above.
$ aws cognito-idp sign-up \
--region us-east-1 \
--client-id 4fb69je3470cat29p0nfm3t27k \
--username admin@example.com \
--password Passw0rd!
Next we’ll verify the user. Replace --user-pool-id
with UserPoolId
from the sst start
output above.
$ aws cognito-idp admin-confirm-sign-up \
--region us-east-1 \
--user-pool-id us-east-1_e8u3sktE1 \
--username admin@example.com
Now we’ll make a request to our private API. Typically, we’ll be using our app to do this. But just to test, we’ll use the AWS API Gateway Test CLI. This makes an authenticated call to our private API using the credentials of the user we just created.
$ npx aws-api-gateway-cli-test \
--username='admin@example.com' \
--password='Passw0rd!' \
--user-pool-id='us-east-1_e8u3sktE1' \
--app-client-id='4fb69je3470cat29p0nfm3t27k' \
--cognito-region='us-east-1' \
--identity-pool-id='us-east-1:d01df859-f416-4dc2-90ac-0c6fc272d197' \
--invoke-url='https://12mflx0e8e.execute-api.us-east-1.amazonaws.com' \
--api-gateway-region='us-east-1' \
--path-template='/private' \
--method='GET'
Make sure to set the options with the ones in your sst start
output.
--user-pool-id
=>UserPoolId
--app-client-id
=>UserPoolClientId
--identity-pool-id
=>IdentityPoolId
--invoke-url
=>ApiEndpoint
--cognito-region
and--api-gateway-region
, the region in yoursst.json
You should now see.
{
status: 200,
statusText: 'OK',
data: 'Hello user!'
}
The above process might seem fairly tedious. But once we integrate it into our frontend app, we’ll be able to use something like AWS Amplify to handle these steps for us.
Making changes
Let’s make a quick change to our private route to print out the caller’s user id.
Replace src/private.js
with the following.
export async function main(event) {
return {
statusCode: 200,
body: `Hello ${event.requestContext.authorizer.iam.cognitoIdentity.identityId}!`,
};
}
We are getting the user id from the event object.
If you make the same authenticated request to the /private
endpoint.
$ npx aws-api-gateway-cli-test \
--username='admin@example.com' \
--password='Passw0rd!' \
--user-pool-id='us-east-1_e8u3sktE1' \
--app-client-id='4fb69je3470cat29p0nfm3t27k' \
--cognito-region='us-east-1' \
--identity-pool-id='us-east-1:d01df859-f416-4dc2-90ac-0c6fc272d197' \
--invoke-url='https://12mflx0e8e.execute-api.us-east-1.amazonaws.com' \
--api-gateway-region='us-east-1' \
--path-template='/private' \
--method='GET'
You should see the user id.
{
status: 200,
statusText: 'OK',
data: 'Hello us-east-1:6f4e594d-a6ca-4a24-b99b-760913a70a31!'
}
Deploying your API
Now that our API is tested and ready to go. Let’s go ahead and deploy it for our users. You’ll recall that we were using a dev
environment, the one specified in your sst.json
.
However, we are going to deploy your API again. But to a different environment, called prod
. This allows us to separate our environments, so when we are working in dev
, it doesn’t break the API for our users.
Run the following in your terminal.
$ npx sst deploy --stage prod
A note on these environments. SST is simply deploying the same app twice using two different stage
names. It prefixes the resources with the stage names to ensure that they don’t thrash.
Cleaning up
Finally, you can remove the resources created in this example using the following command.
$ npx sst remove
And to remove the prod environment.
$ npx sst remove --stage prod
Conclusion
And that’s it! You’ve got a brand new serverless API authenticated with Cognito. A local development environment, to test and make changes. And it’s deployed to production as well, so you can share it with your users. Check out the repo below for the code we used in this example. And leave a comment if you have any questions!
Example repo for reference
github.com/serverless-stack/serverless-stack/tree/master/examples/api-auth-cognitoFor help and discussion
Comments on this exampleMore Examples
APIs
-
REST API
Building a simple REST API.
-
WebSocket API
Building a simple WebSocket API.
-
TypeScript REST API
Building a REST API with TypeScript.
-
Go REST API
Building a REST API with Golang.
-
Custom Domains
Using a custom domain in an API.
Web Apps
Mobile Apps
GraphQL
Databases
-
DynamoDB
Using DynamoDB in a serverless API.
-
MongoDB Atlas
Using MongoDB Atlas in a serverless API.
-
PostgreSQL
Using PostgreSQL and Aurora in a serverless API.
-
CRUD DynamoDB
Building a CRUD API with DynamoDB.
Authentication
Using AWS IAM
-
Facebook Auth
Authenticating a serverless API with Facebook.
-
Google Auth
Authenticating a serverless API with Google.
-
Twitter Auth
Authenticating a serverless API with Twitter.
-
Auth0 IAM
Authenticating a serverless API with Auth0.
Using JWT
-
Cognito JWT
Adding JWT authentication with Cognito.
-
Auth0 JWT
Adding JWT authentication with Auth0.
Async Tasks
-
Cron
A simple serverless Cron job.
-
Queues
A simple queue system with SQS.
-
Pub/Sub
A simple pub/sub system with SNS.
-
Resize Images
Automatically resize images uploaded to S3.
Editors
-
Debug With VS Code
Using VS Code to debug serverless apps.
-
Debug With WebStorm
Using WebStorm to debug serverless apps.
-
Debug With IntelliJ
Using IntelliJ IDEA to debug serverless apps.
Monitoring
Miscellaneous
-
Lambda Layers
Using the chrome-aws-lambda layer to take screenshots.
-
Middy Validator
Use Middy to validate API request and responses.