Aikido Security has identified a vulnerability in GitHub Actions and GitLab CI/CD workflows that allows AI agents to execute malicious instructions, potentially leaking sensitive information. The flaw affects multiple companies and demonstrates how AI prompt injection can compromise software supply chains.

This article outlines how to deploy microservices using Azure Kubernetes Service (AKS) automated through GitHub Actions. It covers the necessary prerequisites, the CI/CD pipeline stages, and best practices for a successful deployment. You’ll learn how to set up the process for building, pushing, and deploying Docker images effectively.

The article discusses how the team automated updates for GitHub Actions runners using Claude AI, enabling seamless management and deployment of updates. This automation significantly reduces manual intervention and streamlines their workflow, enhancing overall efficiency in their development process.

The project provides tools in Go for automated testing against Fastly's WAF simulator, incorporating a CI/CD pipeline with GitHub actions to run tests on code changes. Test cases are structured in YAML format within the test/rules directory, detailing various fields such as identifiers, requests, expected responses, and signals. Users must set up their Fastly NGWAF credentials, run Terraform commands, and check workflow statuses on GitHub to ensure the WAF rules function correctly.

Hosting GitHub Actions runners on HashiCorp Nomad offers a lightweight and scalable alternative to Kubernetes, enabling organizations to run self-hosted runners within their private networks for enhanced security and control. This solution reduces operational costs, simplifies management, and improves deployment speed by utilizing ephemeral runners that minimize resource overhead and ensure clean environments for each job. Additionally, it supports multi-cloud and hybrid deployments, allowing for flexible infrastructure management without vendor lock-in.

Grafana Labs introduced Zizmor, an open source static analysis tool, in their CI/CD pipelines to detect and prevent vulnerabilities in GitHub Actions following a security incident. The tool helps identify unsafe configurations and practices, such as the use of `pull_request_target`, and is part of a broader effort to enhance security across their repositories. Despite facing challenges like GitHub's rate limiting, Grafana is committed to using Zizmor to bolster their defenses against future attacks.

AWS has introduced a new feature that allows for the deployment of AWS Lambda functions directly through GitHub Actions, simplifying the CI/CD process with a declarative YAML configuration. This improvement eliminates the need for manual packaging and configuration steps, enhancing developer experience and security through seamless IAM integration. Users can easily set up a workflow to automatically deploy their functions with minimal effort.

The article discusses the security considerations necessary for using GitHub Actions in CI/CD setups, emphasizing the importance of protecting workflows against potential threats from contributors with write access. It details various attack scenarios, including script injection vulnerabilities, and provides best practices for securing sensitive workflows and managing permissions effectively.

Sysdig's Threat Research Team uncovered significant security vulnerabilities in GitHub Actions workflows across popular open source projects, including those by MITRE and Splunk. Their research revealed how insecure configurations, particularly using pull_request_target, can expose sensitive credentials and allow for exploitation, prompting the team to recommend best practices to enhance CI/CD security.