Secure CDK deployments with IAM permission boundaries

The AWS Cloud Development Kit (CDK) accelerates cloud development by allowing developers to use common programming languages when modelling their applications. To take advantage of this speed, developers need to operate in an environment where permissions and security controls don’t slow things down, and in a tightly controlled environment this is not always the case. Of particular concern is the scenario where a developer has permission to create AWS Identity and Access Management (IAM) entities (such as users or roles), as these could have permissions beyond that of the developer who created them, allowing for an escalation of privileges. This approach is typically controlled through the use of permission boundaries for IAM entities, and in this post you will learn how these boundaries can now be applied more effectively to CDK development – allowing developers to stay secure and move fast.

Time to read
10 minutes

Learning level
Advanced (300)

Services used

AWS Cloud Development Kit (CDK)

AWS Identity and Access Management (IAM)

Applying custom permission boundaries to CDK deployments

When the CDK deploys a solution, it assumes a AWS CloudFormation execution role to perform operations on the user’s behalf. This role is created during the bootstrapping phase by the AWS CDK Command Line Interface (CLI). This role should be configured to represent the maximum set of actions that CloudFormation can perform on the developers behalf, while not compromising any compliance or security goals of the organisation. This can become complicated when developers need to create IAM entities (such as IAM users or roles) and assign permissions to them, as those permissions could be escalated beyond their existing access levels. Taking away the ability to create these entities is one way to solve the problem. However, doing this would be a significant impediment to developers, as they would have to ask an administrator to create them every time. This is made more challenging when you consider that security conscious practices will create individual IAM roles for every individual use case, such as each AWS Lambda Function in a stack. Rather than taking this approach, IAM permission boundaries can help in two ways – first, by ensuring that all actions are within the overlap of the users permissions and the boundary, and second by ensuring that any IAM entities that are created also have the same boundary applied. This blocks the path to privilege escalation without restricting the developer’s ability to create IAM identities. With the latest version of the AWS CLI these boundaries can be applied to the execution role automatically when running the bootstrap command, as well as being added to IAM entities that are created in a CDK stack.

To use a permission boundary in the CDK, first create an IAM policy that will act as the boundary. This should define the maximum set of actions that the CDK application will be able to perform on the developer’s behalf, both during deployment and operation. This step would usually be performed by an administrator who is responsible for the security of the account, ensuring that the appropriate boundaries and controls are enforced. Once created, the name of this policy is provided to the bootstrap command. In the example below, an IAM policy called “developer-policy” is used to demonstrate the command.

cdk bootstrap –custom-permissions-boundary developer-policy

Once this command runs, a new bootstrap stack will be created (or an existing stack will be updated) so that the execution role has this boundary applied to it. Next, you can ensure that any IAM entities that are created will have the same boundaries applied to them. This is done by either using a CDK context variable, or the permissionBoundary attribute on those resources. To explain this in some detail, let’s use a real world scenario and step through an example that shows how this feature can be used to restrict developers from using the AWS Config service.

Installing or upgrading the AWS CDK CLI

Before beginning, ensure that you have the latest version of the AWS CDK CLI tool installed. Follow the instructions in the documentation to complete this. You will need version 2.54.0 or higher to make use of this new feature. To check the version you have installed, run the following command.

cdk –version

Creating the policy

First, let’s begin by creating a new IAM policy. Below is a CloudFormation template that creates a permission policy for use in this example. In this case the AWS CLI can deploy it directly, but this could also be done at scale through a mechanism such as CloudFormation Stack Sets. This template has the following policy statements:

Allow all actions by default – this allows you to deny the specific actions that you choose. You should carefully consider your approach to allow/deny actions when creating your own policies though.
Deny the creation of users or roles unless the “developer-policy” permission boundary is used. Additionally limit the attachment of permissions boundaries on existing entities to only allow “developer-policy” to be used. This prevents the creation or change of an entity that can escalate outside of the policy.
Deny the ability to change the policy itself so that a developer can’t modify the boundary they will operate within.
Deny the ability to remove the boundary from any user or role
Deny any actions against the AWS Config service

Here items 2, 3 and 4 all ensure that the permission boundary works correctly – they are controls that prevent the boundary being removed, tampered with, or bypassed. The real focus of this policy in terms of the example are items 1 and 5 – where you allow everything, except the specific actions that are denied (creating a deny list of actions, rather than an allow list approach).

Type: AWS::IAM::ManagedPolicy
# —– Begin base policy —————
# If permission boundaries do not have an explicit allow
# then the effect is deny
– Sid: ExplicitAllowAll
Action: “*”
Effect: Allow
Resource: “*”
# Default permissions to prevent privilege escalation
– Sid: DenyAccessIfRequiredPermBoundaryIsNotBeingApplied
– iam:CreateUser
– iam:CreateRole
– iam:PutRolePermissionsBoundary
– iam:PutUserPermissionsBoundary
Fn::Sub: arn:${AWS::Partition}:iam::${AWS::AccountId}:policy/developer-policy
Effect: Deny
Resource: “*”
– Sid: DenyPermBoundaryIAMPolicyAlteration
– iam:CreatePolicyVersion
– iam:DeletePolicy
– iam:DeletePolicyVersion
– iam:SetDefaultPolicyVersion
Effect: Deny
Fn::Sub: arn:${AWS::Partition}:iam::${AWS::AccountId}:policy/developer-policy
– Sid: DenyRemovalOfPermBoundaryFromAnyUserOrRole
– iam:DeleteUserPermissionsBoundary
– iam:DeleteRolePermissionsBoundary
Effect: Deny
Resource: “*”
# —– End base policy —————
# — Begin Custom Organization Policy —
– Sid: DenyModifyingConfig
Effect: Deny
Action: config:*
Resource: “*”
# — End Custom Organization Policy —
Version: “2012-10-17”
Description: “Bootstrap Permission Boundary”
ManagedPolicyName: developer-policy
Path: /

Save the above locally as developer-policy.yaml and then you can deploy it with a CloudFormation command in the AWS CLI:

aws cloudformation create-stack –stack-name DeveloperPolicy
–template-body file://developer-policy.yaml

Creating a stack to test the policy

To begin, create a new CDK application that you will use to test and observe the behaviour of the permission boundary. Create a new directory with a TypeScript CDK application in it by executing these commands.

mkdir DevUsers && cd DevUsers
cdk init –language typescript

Once this is done, you should also make sure that your account has a CDK bootstrap stack deployed with the cdk bootstrap command – to start with, do not apply a permission boundary to it, you can add that later an observe how it changes the behaviour of your deployment. Because the bootstrap command is not using the –cloudformation-execution-policies argument, it will default to arn:aws:iam::aws:policy/AdministratorAccess which means that CloudFormation will have full access to the account until the boundary is applied.

cdk bootstrap

Once the command has run, create an AWS Config Rule in your application to be sure that this works without issue before the permission boundary is applied. Open the file lib/dev_users-stack.ts and edit its contents to reflect the sample below.

import * as cdk from ‘aws-cdk-lib’;
import { ManagedRule, ManagedRuleIdentifiers } from ‘aws-cdk-lib/aws-config’;
import { Construct } from “constructs”;

export class DevUsersStack extends cdk.Stack {
constructor(scope: Construct, id: string, props?: cdk.StackProps) {
super(scope, id, props);

new ManagedRule(this, ‘AccessKeysRotated’, {
configRuleName: ‘access-keys-policy’,
identifier: ManagedRuleIdentifiers.ACCESS_KEYS_ROTATED,
inputParameters: {
maxAccessKeyAge: 60, // default is 90 days

Next you can deploy with the CDK CLI using the cdk deploy command, which will succeed (the output below has been truncated to show a summary of the important elements).

❯ cdk deploy
Synthesis time: 3.05s
Deployment time: 23.17s

Stack ARN:

Total time: 26.21s

Before you deploy the permission boundary, remove this stack again with the cdk destroy command.

❯ cdk destroy
Are you sure you want to delete: DevUsersStack (y/n)? y
DevUsersStack: destroying… [1/1]
DevUsersStack: destroyed

Using a permission boundary with the CDK test application

Now apply the permission boundary that you created above and observe the impact it has on the same deployment. To update your booststrap with the permission boundary, re-run the cdk bootstrap command with the new custom-permissions-boundary parameter.

cdk bootstrap –custom-permissions-boundary developer-policy

After this command executes, the CloudFormation execution role will be updated to use that policy as a permission boundary, which based on the deny rule for config:* will cause this same application deployment to fail. Run cdk deploy again to confirm this and observe the error message.

Deployment failed: Error: Stack Deployments Failed: Error: The stack
named DevUsersStack failed creation, it may need to be manually deleted
from the AWS console:
User: arn:aws:sts::123456789012:assumed-role/cdk-hnb659fds-cfn-exec-role-123456789012-ap-southeast-2/AWSCloudFormation
is not authorized to perform: config:PutConfigRule on resource: access-keys-policy with an explicit deny in a
permissions boundary

This shows you that the action was denied specifically due to the use of a permissions boundary, which is what was expected.

Applying permission boundaries to IAM entities automatically

Next let’s explore how the permission boundary can be extended to IAM entities that are created by a CDK application. The concern here is that a developer who is creating a new IAM entity could assign it more permissions than they have themselves – the permission boundary manages this by ensuring that entities can only be created that also have the boundary attached. You can validate this by modifying the stack to deploy a Lambda function that uses a role that doesn’t include the boundary. Open the file lib/dev_users-stack.ts again and edit its contents to reflect the sample below.

import * as cdk from ‘aws-cdk-lib’;
import { PolicyStatement } from “aws-cdk-lib/aws-iam”;
import {
} from “aws-cdk-lib/custom-resources”;
import { Construct } from “constructs”;

export class DevUsersStack extends cdk.Stack {
constructor(scope: Construct, id: string, props?: cdk.StackProps) {
super(scope, id, props);

new AwsCustomResource(this, “Resource”, {
onUpdate: {
service: “ConfigService”,
action: “putConfigRule”,
parameters: {
ConfigRule: {
ConfigRuleName: “SampleRule”,
Source: {
Owner: “AWS”,
SourceIdentifier: “ACCESS_KEYS_ROTATED”,
InputParameters: ‘{“maxAccessKeyAge”:”60″}’,
physicalResourceId: PhysicalResourceId.of(“SampleConfigRule”),
policy: AwsCustomResourcePolicy.fromStatements([
new PolicyStatement({
actions: [“config:*”],
resources: [“*”],

Here the AwsCustomResource is used to provision a Lambda function that will attempt to create a new config rule. This is the same result as the previous stack but in this case the creation of the rule is done by a new IAM role that is created by the CDK construct for you. Attempting to deploy this will result in a failure – run cdk deploy to observe this.

Deployment failed: Error: Stack Deployments Failed: Error: The stack named
DevUsersStack failed creation, it may need to be manually deleted from the AWS
API: iam:CreateRole User: arn:aws:sts::123456789012:assumed-
is not authorized to perform: iam:CreateRole on resource:
with an explicit deny in a permissions boundary

The error message here details that the stack was unable to deploy because the call to iam:CreateRole failed because the boundary wasn’t applied. The CDK now offers a straightforward way to set a default permission boundary on all IAM entities that are created, via the CDK context variable core:permissionsBoundary in the cdk.json file.

“context”: {
“@aws-cdk/core:permissionsBoundary”: {
“name”: “developer-policy”

This approach is useful because now you can import constructs that create IAM entities (such as those found on Construct Hub or out of the box constructs that create default IAM roles) and have the boundary apply to them as well. There are alternative ways to achieve this, such as setting a boundary on specific roles, which can be used in scenarios where this approach does not fit. Make the change to your cdk.json file and run the CDK deploy again. This time the custom resource will attempt to create the config rule using its IAM role instead of the CloudFormation execution role. It is expected that the boundary will also protect this Lambda function in the same way – run cdk deploy again to confirm this. Note that the deployment updates from CloudFormation show that this time the role creation succeeds this time, and a new error message is generated.

Deployment failed: Error: Stack Deployments Failed: Error: The stack named
DevUsersStack failed creation, it may need to be manually deleted from the AWS
Received response status [FAILED] from custom resource. Message returned: User:
is not authorized to perform: config:PutConfigRule on resource: SampleRule with an explicit deny in a permissions boundary

In this error message you can see that the user it refers to is DevUsersStack-AWS679f53fac002430cb0da5b7982bd2287S-84VFVA7OGC9N rather than the CloudFormation execution role. This is the role being used by the custom Lambda function resource, and when it attempts to create the Config rule it is rejected because of the permissions boundary in the same way. Here you can see how the boundary is being applied consistently to all IAM entities that are created in your CDK app, which ensures the administrative controls can be applied consistently to everything a developer does with a minimal amount of overhead.


At this point you can either choose to remove the CDK bootstrap stack if you no longer require it, or remove the permission boundary from the stack. To remove it, delete the CDKToolkit stack from CloudFormation with this AWS CLI command.

aws cloudformation delete-stack –stack-name CDKToolkit

If you want to keep the bootstrap stack, you can remove the boundary by following these steps:

Browse to the CloudFormation page in the AWS console, and select the CDKToolit stack.
Select the ‘Update’ button. Choose “Use Current Template” and then press ‘Next’
On the parameters page, find the value InputPermissionsBoundary which will have developer-policy as the value, and delete the text in this input to leave it blank. Press ‘Next’ and the on the following page, press ‘Next’ again
On the final page, scroll to the bottom and check the box acknowledging that CloudFormation might create IAM resources with custom names, and choose ‘Submit’

With the permission boundary no longer being used, you can now remove the stack that created it as the final step.

aws cloudformation delete-stack –stack-name DeveloperPolicy


Now you can see how IAM permission boundaries can easily be integrated in to CDK development, helping ensure developers have the control they need while administrators can ensure that security is managed in a way that meets the needs of the organisation as well.

With this being understood, there are next steps you can take to further expand on the use of permission boundaries. The CDK Security and Safety Developer Guide document on GitHub outlines these approaches, as well as ways to think about your approach to permissions on deployment. It’s recommended that developers and administrators review this, and work to develop and appropriate approach to permission policies that suit your security goals.

Additionally, the permission boundary concept can be applied in a multi-account model where each Stage has a unique boundary name applied. This can allow for scenarios where a lower-level environment (such as a development or beta environment) has more relaxed permission boundaries that suit troubleshooting and other developer specific actions, but then the higher level environments (such as gamma or production) could have the more restricted permission boundaries to ensure that security risks are more appropriately managed. The mechanism for implement this is defined in the security and safety developer guide also.

About the authors:

Brian Farnhill

Brian Farnhill is a Software Development Engineer at AWS, helping public sector customers in APAC create impactful solutions running in the cloud. His background is in building solutions and helping customers improve DevOps tools and processes. When he isn’t working, you’ll find him either coding for fun or playing online games.

David Turnbull

David Turnbull is a Software Development Engineer at AWS, helping public sector customers in APAC create impactful solutions running in the cloud. He likes to comprehend new programming languages and has used this to stray out of his line. David writes computer simulations for fun.

Team Collaboration with Amazon CodeCatalyst

Amazon CodeCatalyst enables teams to collaborate on features, tasks, bugs, and any other work involved when building software. CodeCatalyst was announced at re:Invent 2022 and is currently in preview.


In a prior post in this series, Using Workflows to Build, Test, and Deploy with Amazon CodeCatalyst, I discussed reading The Unicorn Project, by Gene Kim, and how the main character, Maxine, struggles with a complicated software development lifecycle (SLDC) after joining a new team. Some of the challenges she encounters include:

Continually delivering high-quality updates is complicated and slow
Collaborating efficiently with others is challenging
Managing application environments is increasingly complex
Setting up a new project is a time-consuming chore

In this post, I will focus on the second bullet, and how CodeCatalyst helps you collaborate from anywhere with anyone.


If you would like to follow along with this walkthrough, you will need to:

Have an AWS Builder ID for signing in to CodeCatalyst.
Belong to a CodeCatalyst space and have the Space administrator role assigned to you in that space. For more information, see Creating a space in CodeCatalyst, Managing members of your space, and Space administrator role.
Have an AWS account associated with your space and have the IAM role in that account. For more information about the role and role policy, see Creating a CodeCatalyst service role.


Similar to the prior post, I am going to use the Modern Three-tier Web Application blueprint in this walkthrough. A CodeCatalyst blueprint provides a template for a new project. If you would like to follow along, you can launch the blueprint as described in Creating a project in Amazon CodeCatalyst.  This will deploy the Mythical Mysfits sample application shown in the following image.

Figure 1. The Mythical Mysfits user interface showing header and three Mysfits

For this Walkthrough, let us assume that I need to make a simple change to the application. The legal department would like to add a footer that includes the text “© 2023 Demo Organization.” I will create an issue in CodeCatalyst to track this work and use CodeCatalyst to track the change throughout the entire Software Development Life Cycle (SDLC).

CodeCatalyst organizes projects into Spaces. A space represents your company, department, or group; and contains projects, members, and the associated cloud resources you create in CodeCatalyst. In this walkthrough, my Space currently includes two members, Brian Beach and Panna Shetty, as shown in the following screenshot.  Note that both users are administrators, but CodeCatalyst supports multiple roles. You can read more about roles in members of your space.

Figure 2. The space members configuration page showing two users

To begin, Brian creates a new issue to track the request from legal. He assigns the issue to Panna, but leaves it in the backlog for now. Note that CodeCatalyst supports multiple metadata fields to organize your work. This issue is not impacting users and is relatively simple to fix. Therefore, Brian has categorized it as low priority and estimated the effort as extra small (XS). Brian has also added a label, so all the requests from legal can be tracked together. Note that these metadata fields are customizable. You can read more in configuring issue settings.

Figure 3. Create issue dialog box with name, description and metadata

CodeCatalyst supports rich markdown in the description field. You can read about this in Markdown tips and tricks. In the following screenshot, Brian types “@app.vue” which brings up an inline search for people, issues, and code to help Panna find the relevant bit of code that needs changing later.

Figure 4. Create issue dialog box with type-ahead overlay

When Panna is ready to begin work on the new feature, she moves the issue from the “Backlog“ to ”In progress.“ CodeCatalyst allows users to manage their work using a Kanban style board. Panna can simply drag-and-drop issues on the board to move the issue from one state to another. Given the small team, Brian and Panna use a single board. However, CodeCatalyst allows you to create multiple views filtered by the metadata fields discussed earlier. For example, you might create a label called Sprint-001, and use that to create a board for the sprint.

Figure 5. Kanban board showing to do, in progress and in review columns

Panna creates a new branch for the change called feature_add_copyright and uses the link in the issue description to navigate to the source code repository. This change is so simple that she decides to edit the file in the browser and commits the change. Note that for more complex changes, CodeCatalyst supports Dev Environments. The next post in this series will be dedicated to Dev Environments. For now, you just need to know that a Dev Environment is a cloud-based development environment that you can use to quickly work on the code stored in the source repositories of your project.

Figure 6. Editor with new lines highlighted

Panna also creates a pull request to merge the feature branch in to the main branch. She identifies Brian as a required reviewer. Panna then moves the issue to the “In review” column on the Kanban board so the rest of the team can track the progress. Once Brian reviews the change, he approves and merges the pull request.

Figure 7. Pull request details with title, description, and reviewed assigned

When the pull request is merged, a workflow is configured to run automatically on code changes to build, test, and deploy the change. Note that Workflows were covered in the prior post in this series. Once the workflow is complete, Panna is notified in the team’s Slack channel. You can read more about notifications in working with notifications in CodeCatalyst. She verifies the change in production and moves the issue to the done column on the Kanban board.

Figure 8. Kanban board showing in progress, in review, and done columns

Once the deployment completes, you will see the footer added at the bottom of the page.

Figure 9. The Mythical Mysfits user interface showing footer and three Mysfits

At this point the issue is complete and you have seen how this small team collaborated to progress through the entire software development lifecycle (SDLC).


If you have been following along with this workflow, you should delete the resources you deployed so you do not continue to incur charges. First, delete the two stacks that CDK deployed using the AWS CloudFormation console in the AWS account you associated when you launched the blueprint. These stacks will have names like mysfitsXXXXXWebStack and mysfitsXXXXXAppStack. Second, delete the project from CodeCatalyst by navigating to Project settings and choosing Delete project.


In this post, you learned how CodeCatalyst can help you rapidly collaborate with other developers. I used issues to track feature and bugs, assigned code reviews, and managed pull requests. In future posts I will continue to discuss how CodeCatalyst can address the rest of the challenges Maxine encountered in The Unicorn Project.

About the authors:

Brian Beach

Brian Beach has over 20 years of experience as a Developer and Architect. He is currently a Principal Solutions Architect at Amazon Web Services. He holds a Computer Engineering degree from NYU Poly and an MBA from Rutgers Business School. He is the author of “Pro PowerShell for Amazon Web Services” from Apress. He is a regular author and has spoken at numerous events. Brian lives in North Carolina with his wife and three kids.

Panna Shetty

Panna Shetty is a Sr. Solutions Architect with Amazon Web Services (AWS), working with public sector customers. She enjoys helping customers architect and build scalable and
reliable modern applications using cloud-native technologies.

re:Invent 2022 DevOps and Developer Productivity Playlist

Danielle Kucera, Karun Bakshi, and I were privileged to organize the DevOps and Developer Productivity (DOP) track for re:Invent 2022. For 2022, the DOP track included 58 sessions and nearly 100 speakers.  If you weren’t able to attend, I have compiled a list of the on-demand sessions for you below.

Leadership Sessions

Delighting developers: Builder experience at AWS Adam Seligman, Vice President of Developer Experience, and Emily Freeman, Head of Community Development, share the latest AWS tools and experiences for teams developing in the cloud. Adam recaps the latest launches and demos how key services can integrate to accelerate developer productivity.

Amazon CodeCatalyst

Amazon CodeCatalyst, announced during Dr. Werner Vogels Keynote, is a unified software development service that makes it faster to build and deliver on AWS.

Introducing Amazon CodeCatalyst – Harry Mower, Director of DevOps Services, and Doug Clauson, Product Manager, provide an overview of Amazon CodeCatalyst. CodeCatalyst provides one place where you can plan work, collaborate on code, and build, test, and deploy applications with nearly continuous integration/continuous delivery (CI/CD) tools.

Deep dive on CodeCatalyst Workspaces – Tmir Karia, Sr. Product Manager, and Rahul Gulati, Sr. Product Manager,  discuss how Amazon CodeCatalyst Workspaces decreases the time you spend creating and maintaining a local development environment and allows you to quickly set up a cloud development workspace, switch between projects, and replicate the development workspace configuration across team members.


AWS Well-Architected best practices for DevOps on AWS Elamaran Shanmugam, Sr. Container Specialist, and Deval Perikh, Sr. Enterprise Solutions Architect, discuss the components required to align your DevOps practices to the pillars of the AWS Well-Architected Framework.

Best practices for securing your software delivery lifecycle Jams Bland, Principal Solutions Architect, and Curtis Rissi, Principal Solutions Architect, discus ways you can secure your CI/CD pipeline on AWS. Review topics like security of the pipeline versus security in the pipeline, ways to incorporate security checkpoints across various pipeline stages, security event management, and aggregating vulnerability findings into a single pane of glass.

Build it & run it: Streamline your DevOps capabilities with machine learning Rafael Ramos, Shivansh Singh, and Jared Reimer discuss how to use machine learning–powered tools like Amazon CodeWhisperer, Amazon CodeGuru, and Amazon DevOps Guru to boost your applications’ availability and write software faster and more reliably.

Infrastructure as Code

AWS infrastructure as code: A year in review  Tatiana Cooke, Principal Product Manager, and Ben Perak, Principal Product Manage, discuss the new features and improvements for AWS infrastructure as code with AWS CloudFormation and AWS CDK.

How to reuse patterns when developing infrastructure as code Ryan Bachman, Ethan Rucinski, and Ravi Palakodeti explore AWS Cloud Development Kit (AWS CDK) constructs and AWS CloudFormation modules and how they make it easier to build applications on AWS.

Governance and security with infrastructure as code David Hessler, Senior DevOps Consultant, and Eric Beard, Senior Solutions Architect, discuss how to use AWS CloudFormation and the AWS CDK to deploy cloud applications in regulated environments while enforcing security controls.

Developer Productivity

Building on AWS with AWS tools, services, and SDKs Kyle Thomson, Senior Software Development Engineer and Deval Parikh, Senior Solutions Architect, discuss the ways developers can set up secure development environments and use their favorite IDEs to interact with, and deploy to, the AWS Cloud.

The Amazon Builders’ Library: 25 years of operational excellence at Amazon Colm MacCarthaigh, Distinguished Engineer, and David Yanacek, Sr. Principal Engineer, discuss how Amazon practices have changed and improved over time and what we’ve learned as builders and as operators.

Sustainability in the cloud with Rust and AWS Graviton Emil Lerch, Principal DevOps Specialist, and Esteban Kuber, Principal Engineer, discuss the benefits of Rust and AWS Graviton that can reduce energy consumption and increase productivity.



About the author:

Brian Beach

Brian Beach has over 20 years of experience as a Developer and Architect. He is currently a Principal Solutions Architect at Amazon Web Services. He holds a Computer Engineering degree from NYU Poly and an MBA from Rutgers Business School. He is the author of “Pro PowerShell for Amazon Web Services” from Apress. He is a regular author and has spoken at numerous events. Brian lives in North Carolina with his wife and three kids.

Setting up a secure CI/CD pipeline in a private Amazon Virtual Private Cloud with no public internet access

With the rise of the cloud and increased security awareness, the use of private Amazon VPCs with no public internet access also expanded rapidly. This setup is recommended to make sure of proper security through isolation. The isolation requirement also applies to code pipelines, in which developers deploy their application modules, software packages, and other dependencies and bundles throughout the development lifecycle. This is done without having to push larger bundles from the developer space to the staging space or the target environment. Furthermore, AWS CodeArtifact is used as an artifact management service that will help organizations of any size to securely store, publish, and share software packages used in their software development process.

We’ll walk through the steps required to build a secure, private continuous integration/continuous development (CI/CD) pipeline with no public internet access while maintaining log retention in Amazon CloudWatch. We’ll utilize AWS CodeCommit for source, CodeArtifact for the Modules and software packages, and Amazon Simple Storage Service (Amazon S3) as artifact storage.


The prerequisites for following along with this post include:

An AWS Account
A Virtual Private Cloud (Amazon VPC)
A CI/CD pipeline – This can be CodePipeline, Jenkins or any CI/CD tool you want to integrate CodeArtifact with, we will use CodePipeline in our walkthrough here.

Solution walkthrough

The main service we’ll focus on is CodeArtifact, a fully managed artifact repository service that makes it easy for organizations of any size to securely store, publish, and share software packages used in their software development process. CodeArtifact works with commonly used package managers and build tools, such as Maven and Gradle (Java), npm and yarn (JavaScript), pip and twine (Python), or NuGet (.NET).

Users push code to CodeCommit, CodePipeline will detect the change and start the pipeline, in CodeBuild the build stage will utilize the private endpoints and download the software packages needed without the need to go over the internet.

The preceding diagram shows how the requests remain private within the VPC and won’t go through the Internet gateway, by going from CodeBuild over the private endpoint to CodeArtifact service, all within the private subnet.

The requests will use the following VPC endpoints to connect to these AWS services:

CloudWatch Logs endpoint (for CodeBuild to put logs in CloudWatch)
CodeArtifact endpoints

AWS Security Token Service (AWS STS) endpoint
Amazon Simple Storage Service (Amazon S3) endpoint


Create a CodeCommit Repository:

Navigate to your CodeCommit Console then click on Create repository

Figure 2. Screenshot: Create repository button.

Type in name for the repository then click Create

Figure 3. Screenshot: Repository setting with name shown as “Private” and empty Description.

Scroll down and click Create file

Figure 4. Create file button.

Copy the example buildspec.yml file and paste it to the editor

Example buildspec.yml file:

version: 0.2
nodejs: 16

– ACCT=`aws sts get-caller-identity –region ${AWS_REGION} –query Account –output text`
– aws codeartifact login –tool npm –repository Private –domain private –domain-owner ${ACCT}
– npm install
– node index.js

Name the file buildspec.yml, type in your name and your email address then Commit changes

Figure 5. Screenshot: Create file page.

Create CodeArtifact

Navigate to your CodeArtifact Console then click on Create repository
Give it a name and select npm-store as public upsteam repository

Figure 6. Screenshot: Create repository page with Repository name “Private”.

For the Domain Select this AWS account and enter a domain name

Figure 7. Screenshot: Select domain page.

Click Next then Create repository

Figure 8. Screenshot: Create repository review page.

Create a CI/CD using CodePipeline

Navigate to your CodePipeline Console then click on Create pipeline

Figure 9. Screenshot: Create pipeline button.

Type a name, leave the Service role as “New service role” and click next

Figure 10. Screenshot: Choose pipeline setting page with pipeline name “Private”.

Select AWS CodeCommit as your Source provider
Then choose the CodeCommit repository you created earlier and for branch select main then click Next

Figure 11. Screenshot: Create pipeline add source stage.

For the Build Stage, Choose AWS CodeBuild as the build provider, then click Create Project

Figure 12. Screenshot: Create pipeline add build stage.

This will open new window to create the new Project, Give this project a name

Figure 13. Screenshot: Create pipeline create build project window.

 Scroll down to the Environment section: select pick Managed image,
For Operating system select “Amazon Linux 2”,
Runtime “Standard” and
For Image select the aws/codebuild/amazonlinux2-x86+64-standard:4.0
For the Image version: Always use the latest image for this runtime version
Select Linux for the Environment type
Leave the Privileged option unchecked and set Service Role to “New service role”

Figure 14. Screenshot: Create pipeline create build project, setting up environment window.

Expand Additional configurations and scroll down to the VPC section, select the desired VPC, your Subnets (we recommend selecting multiple AZs, to ensure high availability), and Security Group (the security group rules must allow resources that will use the VPC endpoint to communicate with the AWS service to communicate with the endpoint network interface, default VPC security group will be used here as an example)

Figure 15. Screenshot: Create pipeline create build project networking window.

Scroll down to the Buildspec and select “Use a buildspec file” and type “buildspec.yml” for the Buildspec name

Figure 16. Screenshot: Create pipeline create build project buildspec window.

Select the CloudWatch logs option you can leave the group name and stream empty this will let the service use the default values and click Continue to CodePipeline

Figure 17. Screenshot: Create pipeline create build project logs window.

This will create the new CodeBuild Project, update the CodePipeline page, now you can click Next

Figure 18. Screenshot: Create pipeline add build stage window.

 Since we are not deploying this to any environment, you can skip the deploy stage and click “Skip deploy stage”

Figure 19. Screenshot: Create pipeline add deploy stage.

Figure 20. Screenshot: Create pipeline skip deployment stage confirmation.

After you get the popup click skip again you’ll see the review page, scroll all the way down and click Create Pipeline

Create a VPC endpoint for Amazon CloudWatch Logs. This will enable CodeBuild to send execution logs to CloudWatch:

Navigate to your VPC console, and from the navigation menu on the left select “Endpoints”.

Figure 21. Screenshot: VPC endpoint.

 click Create endpoint Button.

Figure 22. Screenshot: Create endpoint.

For service Category, select “AWS Services”. You can set a name for the new endpoint, and make sure to use something descriptive.

Figure 23. Screenshot: Create endpoint page.

From the list of services, search for the endpoint by typing logs in the search bar and selecting the one with
This walkthrough can be done in any region that supports the services. I am going to be using us-west-2, please select the appropriate region for your workload.

Figure 24. Screenshot: create endpoint select services with selected.

Select the VPC that you want the endpoint to be associated with, and make sure that the Enable DNS name option is checked under additional settings.

Figure 25. Screenshot: create endpoint VPC setting shows VPC selected.

Select the Subnets where you want the endpoint to be associated, and you can leave the security group as default and the policy as empty.

Figure 26. Screenshot: create endpoint subnet setting shows 2 subnet selected and default security group selected.

Select Create Endpoint.

Figure 27. Screenshot: create endpoint button.

Create a VPC endpoint for CodeArtifact. At the time of writing this article, CodeArifact has two endpoints: one is for API operations like service level operations and authentication, and the other is for using the service such as getting modules for our code. We’ll need both endpoints to automate working with CodeArtifact. Therefore, we’ll create both endpoints with DNS enabled.

In addition, we’ll need AWS Security Token Service (AWS STS) endpoint for get-caller-identity API call:

Follow steps a-c from the steps that were used from the creating the Logs endpoint above.

a. From the list of services, you can search for the endpoint by typing codeartifact in the search bar and selecting the one with

Figure 28. Screenshot: create endpoint select services with selected.

Follow steps e-g from Part 4.

Then, repeat the same for service.

Figure 29. Screenshot: create endpoint select services with selected.

Enable a VPC endpoint for AWS STS:

Follow steps a-c from Part 4

a. From the list of services you can search for the endpoint by typing sts in the search bar and selecting the one with

Figure 30.Screenshot: create endpoint select services with selected.

Then follow steps e-g from Part 4.

Create a VPC endpoint for S3:

Follow steps a-c from Part 4

a. From the list of services you can search for the endpoint by typing sts in the search bar and selecting the one with, select the one with type of Gateway

Then select your VPC, and select the route tables for your subnets, this will auto update the route table with the new S3 endpoint.

Figure 31. Screenshot: create endpoint select services with selected.

Now we have all of the endpoints set. The last step is to update your pipeline to point at the CodeArtifact repository when pulling your code dependencies. I’ll use CodeBuild buildspec.yml as an example here.

Make sure that your CodeBuild AWS Identity and Access Management (IAM) role has the permissions to perform STS and CodeArtifact actions.

Navigate to IAM console and click Roles from the left navigation menu, then search for your IAM role name, in our case since we selected “New service role” option in step 2.k was created with the name “codebuild-Private-service-role” (codebuild-<BUILD PROJECT NAME>-service-role)

Figure 32. Screenshot: IAM roles with codebuild-Private-service-role role shown in search.

From the Add permissions menu, click on Create inline policy

Search for STS in the services then select STS

Figure 34. Screenshot: IAM visual editor with sts shown in search.

Search for “GetCallerIdentity” and select the action

Figure 35. Screenshot: IAM visual editor with GetCallerIdentity in search and action selected.

Repeat the same with “GetServiceBearerToken”

Figure 36. Screenshot: IAM visual editor with GetServiceBearerToken in search and action selected.

Click on Review, add a name then click on Create policy

Figure 37. Screenshot: Review page and Create policy button.

You should see the new inline policy added to the list

Figure 38. Screenshot: shows the new in-line policy in the list.

For CodeArtifact actions we will do the same on that role, click on Create inline policy

Figure 39. Screenshot: attach policies.

Search for CodeArtifact in the services then select CodeArtifact

Figure 40. Screenshot: select service with CodeArtifact in search.

Search for “GetAuthorizationToken” in actions and select that action in the check box

Figure 41. CodeArtifact: with GetAuthorizationToken in search.

Repeat for “GetRepositoryEndpoint” and “ReadFromRepository”

Click on Resources to fix the 2 warnings, then click on Add ARN on the first one “Specify domain resource ARN for the GetAuthorizationToken action.”

Figure 42. Screenshot: with all selected filed and 2 warnings.

You’ll get a pop up with fields for Region, Account and Domain name, enter your region, your account number, and the domain name, we used “private” when we created our domain earlier.

Figure 43. Screenshot: Add ARN page.

Then click Add

Repeat the same process for “Specify repository resource ARN for the ReadFromRepository and 1 more”, and this time we will provide Region, Account ID, Domain name and Repository name, we used “Private” for the repository we created earlier and “private” for domain

Figure 44. Screenshot: add ARN page.

Note it is best practice to specify the resource we are targeting, we can use the checkbox for “Any” but we want to narrow the scope of our IAM role best we can.

Navigate to CodeCommit then click on the repo you created earlier in step1

Figure 45. Screenshot: CodeCommit repo.

Click on Add file dropdown, then Create file button

Paste the following in the editor space:

“dependencies”: {
“mathjs”: “^11.2.0”

Name the file “package.json”

Add your name and email, and optional commit message

Repeat this process for “index.js” and paste the following in the editor space:

const { sqrt } = require(‘mathjs’)

Figure 46. Screenshot: CodeCommit Commit changes button.

This will force the pipeline to kick off and start building the application

Figure 47. Screenshot: CodePipeline.

This is a very simple application that gets the square root of 49 and log it to the screen, if you click on the Details link from the pipeline build stage, you’ll see the output of running the NodeJS application, the logs are stored in CloudWatch and you can navigate there by clicking on the link the View entire log “Showing the last xx lines of the build log. View entire log”

Figure 48. Screenshot: Showing the last 54 lines of the build log. View entire log.

We used npm example in the buildspec.yml above, Similar setup will be used for pip and twine,

For Maven, Gradle, and NuGet, you must set Environment variables and change your settings.xml and build.gradle, as well as install the plugin for your IDE. For more information, see here.


Navigate to VPC endpoint from the AWS console and delete the endpoints that you created.

Navigate to CodePipeline and delete the Pipeline you created.

Navigate to CodeBuild and delete the Build Project created.

Navigate to CodeCommit and delete the Repository you created.

Navigate to CodeArtifact and delete the Repository and the domain you created.

Navigate to IAM and delete the Roles created:

For CodeBuild: codebuild-<Build Project Name>-service-role

For CodePipeline: AWSCodePipelineServiceRole-<Region>-<Project Name>


In this post, we deployed a full CI/CD pipeline with CodePipeline orchestrating CodeBuild to build and test a small NodeJS application, using CodeArtifact to download the application code dependencies. All without going to the public internet and maintaining the logs in CloudWatch.

About the author:

MJ Kubba

MJ Kubba is a Solutions Architect who enjoys working with public sector customers to build solutions that meet their business needs. MJ has over 15 years of experience designing and implementing software solutions. He has a keen passion for DevOps and cultural transformation.