GitHub Actions to Securely Publish NPM Packages

GitHub Actions are growing in popularity ever since GitHub announced general availability for all developers and repositories on the GitHub platform. Fueled with some rate limits we’re seeing in the ecosystem—such as new billing and rate limits for open source from Travis CI—will further drive developers to migrate their software automations to GitHub Actions.

In this article, I’d like to show you how I’m using GitHub Actions to publish npm packages that I maintain in my open source projects. If you’re following the GitHub flow which consists of GitHub pull requests workflows, then this will further unify your experience around GitHub workflows for your teams and community contributors in your project.

What is GitHub Actions?

GitHub Actions is a technology developed by GitHub to provide developers with a way to automate their workflows around continuous integration—helping them build, deploy, schedule recurring tasks, and more. GitHub Actions is natively available and integrated into GitHub repositories and features many reusable workflows from community contributors, such as publishing npm packages, publishing docker images, running security tests, and more.

How do actions work in GitHub?

GitHub’s cloud infrastructure is running GitHub Actions for users by creating a workflow file in a GitHub repository directory named .github/workflows, describing the trigger, schedule for the job, and the actions the job should take using YAML.

What is a GitHub workflow?

A GitHub workflow is a set of jobs that would run based on a trigger or a cron-based schedule. A job consists of one or more steps that make up an automated workflow.

Setting up a Node.js project for GitHub Actions

Let’s add an initial GitHub Actions automation to a Node.js project. The GitHub Actions job will install all required npm packages, run tests, and eventually publish our project as an npm package that users can consume.

Our npm package is going to be a Command Line Interface (CLI) for you to browse the amazing list of talks from SnykCon 2020—Snyk’s first-ever global security event that took place in 2020.

The complete project is hosted on GitHub and here is a sneak peek at how the npm package looks like:

A complete GitHub CI workflow starts off with creating the following GitHub Action file at the root of the repository path: .github/workflows/main.yml

name: main

on: [push]

jobs:
  build:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        node-version: [12.x, 14.x]
    steps:
      - uses: actions/checkout@v2
      - name: Build on Node.js ${{ matrix.node-version }}
        uses: actions/setup-node@v1
        with:
          node-version: ${{ matrix.node-version }}
      - run: npm ci
      - run: npm run build --if-present
        name: Build
      - run: npm test
        env:
          CI: true

The above is very much a stock template for a build and test process of Node.js projects, doing the following:

1. The trigger is on every push, to every branch.
2. We run this action on Node.js versions 12.x and 14.x to ensure compatibility across both Node.js LTS versions.
3. The job then runs several steps that check out the git codebase, runs a secure and deterministic npm install phase, continues to run a build step if present, and finally runs tests.

As you push new commits to the repository, whether you are following a GitHub flow, or a GitHub pull request workflow, you’ll see new jobs queuing up in the repository’s Actions tab. Like the following, that shows our tests running for npm package snykcon:

Publishing npm packages with GitHub Actions

If all tests pass, and this automation workflow runs in our main branch, we could automate releasing new versions altogether!

When it comes to releasing packages, it’s a good idea to consider the security implications that concern such actions. Let’s review a few of them:

1. Malicious code: If a vulnerability is intentionally added by someone as part of the code contribution and you missed it, an automatic release when merging to your main branch means it will be available to users immediately. If you manually release immediately, there’s no difference either—however, the more time passes between merging pull requests and releases, it gives more time for the community to scrutinize and help flesh out such issues.
2. Vulnerable dependency: If a malicious person adds a dependency with known public vulnerability then this dependency will trickle into your users as it gets pulled during the install process. Using a free tool like Snyk to connect to your git repositories and add a status check to prevent you from releasing with vulnerable versions of dependencies, solves exactly this problem.
3. Malicious dependency through a lockfile: Have you considered what would happen if a contribution to update package.json dependencies would actually inject a malicious package into the lockfile, which nobody takes the time to review anyway, right? I wrote about why npm lockfiles can be a security blindspot for injecting malicious modules so make sure you deep dive into this if you hadn’t considered this attack vector.
4. Stealing your npm token: In the past, quite a few attempts of malicious packages circled around the notion of stealing a person’s npm token or other sensitive information that exists in environment variables.

The above isn’t meant to be a comprehensive list of security concerns, but it definitely highlights quite a few that we should be aware of.

Speaking of security highlights, do you find that list of possible security concerns an interesting and educational read? It’s a quick threat modeling process, which you can practice more regularly when you build out features. We wrote about it in our DevSecOps Process and there’s a good talk from SnykCon by Alyssa Miller on User Story Threat Modeling: It’s the DevSecOps Way. Check them out!

Let’s go back to our list and focus on the last security concern mentioned—stealing your npm token. Since this article is all about publishing npm packages, it means we need to make an npm token available to the GitHub Actions workflow and this has historically been frowned upon for the following reasons:

• npm capabilities: historically, releasing npm packages using an npm token, required your npm user to disable two-factor authentication. It’s not anymore and we’re going to learn how!
• Stealing a token from malicious packages install: if you make the npm token available in your CI as environment variables, then malicious packages that exist in your package dependency tree (beyond your own direct dependencies) may have access to it during, for example, the npm install process, which by default allows packages to run any arbitrary command.

#github #npm #node #javascript #developer

What is GEEK

Buddha Community

Publish JS Package to NPM or Github

https://www.youtube.com/playlist?list=PLT5Jhb7lgSBOnwyukLUZb6X-WnliySCUL

#github #publish js packag #js package #npm

Best Custom Web & Mobile App Development Company

Everything around us has become smart, like smart infrastructures, smart cities, autonomous vehicles, to name a few. The innovation of smart devices makes it possible to achieve these heights in science and technology. But, data is vulnerable, there is a risk of attack by cybercriminals. To get started, let’s know about IoT devices.

What are IoT devices?

The Internet Of Things(IoT) is a system that interrelates computer devices like sensors, software, and actuators, digital machines, etc. They are linked together with particular objects that work through the internet and transfer data over devices without humans interference.

Famous examples are Amazon Alexa, Apple SIRI, Interconnected baby monitors, video doorbells, and smart thermostats.

How could your IoT devices be vulnerable?

When technologies grow and evolve, risks are also on the high stakes. Ransomware attacks are on the continuous increase; securing data has become the top priority.

When you think your smart home won’t fudge a thing against cybercriminals, you should also know that they are vulnerable. When cybercriminals access our smart voice speakers like Amazon Alexa or Apple Siri, it becomes easy for them to steal your data.

Cybersecurity report 2020 says popular hacking forums expose 770 million email addresses and 21 million unique passwords, 620 million accounts have been compromised from 16 hacked websites.

The attacks are likely to increase every year. To help you secure your data of IoT devices, here are some best tips you can implement.

Tips to secure your IoT devices

1. Change Default Router Name

Your router has the default name of make and model. When we stick with the manufacturer name, attackers can quickly identify our make and model. So give the router name different from your addresses, without giving away personal information.

2. Know your connected network and connected devices

If your devices are connected to the internet, these connections are vulnerable to cyber attacks when your devices don’t have the proper security. Almost every web interface is equipped with multiple devices, so it’s hard to track the device. But, it’s crucial to stay aware of them.

3. Change default usernames and passwords

When we use the default usernames and passwords, it is attackable. Because the cybercriminals possibly know the default passwords come with IoT devices. So use strong passwords to access our IoT devices.

4. Manage strong, Unique passwords for your IoT devices and accounts

Use strong or unique passwords that are easily assumed, such as ‘123456’ or ‘password1234’ to protect your accounts. Give strong and complex passwords formed by combinations of alphabets, numeric, and not easily bypassed symbols.

Also, change passwords for multiple accounts and change them regularly to avoid attacks. We can also set several attempts to wrong passwords to set locking the account to safeguard from the hackers.

5. Do not use Public WI-FI Networks

Are you try to keep an eye on your IoT devices through your mobile devices in different locations. I recommend you not to use the public WI-FI network to access them. Because they are easily accessible through for everyone, you are still in a hurry to access, use VPN that gives them protection against cyber-attacks, giving them privacy and security features, for example, using Express VPN.

6. Establish firewalls to discover the vulnerabilities

There are software and firewalls like intrusion detection system/intrusion prevention system in the market. This will be useful to screen and analyze the wire traffic of a network. You can identify the security weakness by the firewall scanners within the network structure. Use these firewalls to get rid of unwanted security issues and vulnerabilities.

7. Reconfigure your device settings

Every smart device comes with the insecure default settings, and sometimes we are not able to change these default settings configurations. These conditions need to be assessed and need to reconfigure the default settings.

8. Authenticate the IoT applications

Nowadays, every smart app offers authentication to secure the accounts. There are many types of authentication methods like single-factor authentication, two-step authentication, and multi-factor authentication. Use any one of these to send a one time password (OTP) to verify the user who logs in the smart device to keep our accounts from falling into the wrong hands.

9. Update the device software up to date

Every smart device manufacturer releases updates to fix bugs in their software. These security patches help us to improve our protection of the device. Also, update the software on the smartphone, which we are used to monitoring the IoT devices to avoid vulnerabilities.

10. Track the smartphones and keep them safe

When we connect the smart home to the smartphone and control them via smartphone, you need to keep them safe. If you miss the phone almost, every personal information is at risk to the cybercriminals. But sometimes it happens by accident, makes sure that you can clear all the data remotely.

However, securing smart devices is essential in the world of data. There are still cybercriminals bypassing the securities. So make sure to do the safety measures to avoid our accounts falling out into the wrong hands. I hope these steps will help you all to secure your IoT devices.

If you have any, feel free to share them in the comments! I’d love to know them.

Are you looking for more? Subscribe to weekly newsletters that can help your stay updated IoT application developments.

#iot #enterprise iot security #how iot can be used to enhance security #how to improve iot security #how to protect iot devices from hackers #how to secure iot devices #iot security #iot security devices #iot security offerings #iot security technologies iot security plus #iot vulnerable devices #risk based iot security program

How to Set Up GitHub Actions to Publish a Monorepo to NPM

In my last post, I showed you how to create, test, and build a monorepo repository. In this article, I will show you how to automate the publishing of your monorepo to NPM using GitHub Actions.

What are GitHub Actions?

GitHub Actions allow for automating workflows based on repository events such as push, issue creation, or the creation of a new release.

Workflows are composed of jobs, which run concurrently by default. Each job should represent a separate part of your workflow described using steps.

For the propose of this article, we will have one job that will describe what steps must be followed to publish our package.

#publishing #github-actions #javascript #monorepo #npm

GitHub Actions to Securely Publish NPM Packages

GitHub Actions are growing in popularity ever since GitHub announced general availability for all developers and repositories on the GitHub platform. Fueled with some rate limits we’re seeing in the ecosystem—such as new billing and rate limits for open source from Travis CI—will further drive developers to migrate their software automations to GitHub Actions.

In this article, I’d like to show you how I’m using GitHub Actions to publish npm packages that I maintain in my open source projects. If you’re following the GitHub flow which consists of GitHub pull requests workflows, then this will further unify your experience around GitHub workflows for your teams and community contributors in your project.

What is GitHub Actions?

GitHub Actions is a technology developed by GitHub to provide developers with a way to automate their workflows around continuous integration—helping them build, deploy, schedule recurring tasks, and more. GitHub Actions is natively available and integrated into GitHub repositories and features many reusable workflows from community contributors, such as publishing npm packages, publishing docker images, running security tests, and more.

How do actions work in GitHub?

GitHub’s cloud infrastructure is running GitHub Actions for users by creating a workflow file in a GitHub repository directory named .github/workflows, describing the trigger, schedule for the job, and the actions the job should take using YAML.

What is a GitHub workflow?

A GitHub workflow is a set of jobs that would run based on a trigger or a cron-based schedule. A job consists of one or more steps that make up an automated workflow.

Setting up a Node.js project for GitHub Actions

Let’s add an initial GitHub Actions automation to a Node.js project. The GitHub Actions job will install all required npm packages, run tests, and eventually publish our project as an npm package that users can consume.

Our npm package is going to be a Command Line Interface (CLI) for you to browse the amazing list of talks from SnykCon 2020—Snyk’s first-ever global security event that took place in 2020.

The complete project is hosted on GitHub and here is a sneak peek at how the npm package looks like:

A complete GitHub CI workflow starts off with creating the following GitHub Action file at the root of the repository path: .github/workflows/main.yml

name: main

on: [push]

jobs:
  build:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        node-version: [12.x, 14.x]
    steps:
      - uses: actions/checkout@v2
      - name: Build on Node.js ${{ matrix.node-version }}
        uses: actions/setup-node@v1
        with:
          node-version: ${{ matrix.node-version }}
      - run: npm ci
      - run: npm run build --if-present
        name: Build
      - run: npm test
        env:
          CI: true

The above is very much a stock template for a build and test process of Node.js projects, doing the following:

1. The trigger is on every push, to every branch.
2. We run this action on Node.js versions 12.x and 14.x to ensure compatibility across both Node.js LTS versions.
3. The job then runs several steps that check out the git codebase, runs a secure and deterministic npm install phase, continues to run a build step if present, and finally runs tests.

As you push new commits to the repository, whether you are following a GitHub flow, or a GitHub pull request workflow, you’ll see new jobs queuing up in the repository’s Actions tab. Like the following, that shows our tests running for npm package snykcon:

Publishing npm packages with GitHub Actions

If all tests pass, and this automation workflow runs in our main branch, we could automate releasing new versions altogether!

When it comes to releasing packages, it’s a good idea to consider the security implications that concern such actions. Let’s review a few of them:

1. Malicious code: If a vulnerability is intentionally added by someone as part of the code contribution and you missed it, an automatic release when merging to your main branch means it will be available to users immediately. If you manually release immediately, there’s no difference either—however, the more time passes between merging pull requests and releases, it gives more time for the community to scrutinize and help flesh out such issues.
2. Vulnerable dependency: If a malicious person adds a dependency with known public vulnerability then this dependency will trickle into your users as it gets pulled during the install process. Using a free tool like Snyk to connect to your git repositories and add a status check to prevent you from releasing with vulnerable versions of dependencies, solves exactly this problem.
3. Malicious dependency through a lockfile: Have you considered what would happen if a contribution to update package.json dependencies would actually inject a malicious package into the lockfile, which nobody takes the time to review anyway, right? I wrote about why npm lockfiles can be a security blindspot for injecting malicious modules so make sure you deep dive into this if you hadn’t considered this attack vector.
4. Stealing your npm token: In the past, quite a few attempts of malicious packages circled around the notion of stealing a person’s npm token or other sensitive information that exists in environment variables.

The above isn’t meant to be a comprehensive list of security concerns, but it definitely highlights quite a few that we should be aware of.

Speaking of security highlights, do you find that list of possible security concerns an interesting and educational read? It’s a quick threat modeling process, which you can practice more regularly when you build out features. We wrote about it in our DevSecOps Process and there’s a good talk from SnykCon by Alyssa Miller on User Story Threat Modeling: It’s the DevSecOps Way. Check them out!

Let’s go back to our list and focus on the last security concern mentioned—stealing your npm token. Since this article is all about publishing npm packages, it means we need to make an npm token available to the GitHub Actions workflow and this has historically been frowned upon for the following reasons:

• npm capabilities: historically, releasing npm packages using an npm token, required your npm user to disable two-factor authentication. It’s not anymore and we’re going to learn how!
• Stealing a token from malicious packages install: if you make the npm token available in your CI as environment variables, then malicious packages that exist in your package dependency tree (beyond your own direct dependencies) may have access to it during, for example, the npm install process, which by default allows packages to run any arbitrary command.

#github #npm #node #javascript #developer

Publish NuGet Packages Using GitHub Actions

Unbelievable! One of the oldest NetLicensing Client libraries for C# (with the first GitHub push made on Oct 2, 2013) was not available in one of the essential package managers for .NET

Better late than never — so we decided to change this status quo, and today want to share with you our this walkthrough on how we published NetLicensing C## Client library to NuGet repository using GitHub Actions.

Background Info — GitHub Flow

Before moving forward, just a note about GitHub Flow workflow adopted for Labs64 projects hosted at GitHub.

“Pull Request” or “PR” is a very useful feature of the GitHub version control system and allowing efficient feature and bugfix development with the GitHub Flow.

The below diagram showing GitHub Flow adopted for Labs64 projects:

Any new feature or defect fix implementation is being done only in a dedicated feature branch. When branch implementation is ready to be integrated into the master branch, a Pull Request gets created. Using this PR, team members, working on a particular feature/enhancement/bug fix, can get feedback from other team members along the way.

This feedback is being used to make further changes and commits to the branch before finally merging the changes back up to the 'master' branch.

For the above GitHub Flow following workflows will be defined:

• CI — build and test push commits at Pull Request branches and master
• Release — package and publish C## library after successful CI run on 'master'

#tutorial #devops #c# #github #csharp #nuget #continious integration #github actions #nuget packages