Mass GitHub Account Hijacking Spreads CI Secret-Stealing Malware Across Thousands of Repos
Aikido Security reports that an attacker gained control of 4,837 GitHub accounts and used their write access to plant a secret-stealing payload inside more than 1,806 repositories, beginning on a Tuesday this week. By leveraging accounts that already had commit or workflow permissions, the attacker was able to insert malicious code directly into continuous integration (CI) pipelines without needing to breach GitHub's own infrastructure.
CI/CD systems are a prime target for this kind of attack because they routinely handle sensitive secrets such as API keys, cloud credentials, signing certificates, and deployment tokens. A stealer embedded in a build pipeline can silently harvest these values from environment variables or runtime memory and exfiltrate them to an attacker-controlled server, often before any human reviewer notices the change. Because CI jobs run with elevated trust and broad access to downstream systems, a single compromised pipeline can cascade into access across cloud accounts, package registries, and production infrastructure far beyond the original repository.
The scale of this incident — thousands of hijacked accounts feeding malicious code into nearly two thousand repositories — illustrates how credential compromise at the developer-account level can quickly become a software-supply-chain problem. Attackers increasingly favor this approach over attacking build systems directly, since stolen or reused credentials, phished tokens, or leaked personal access tokens provide a quieter and more scalable route to inserting malicious commits that blend into legitimate development activity.
Organizations should treat this as a reminder to audit GitHub account security at scale: enforce mandatory multi-factor authentication for all contributors, rotate and scope personal access tokens tightly, and monitor for unusual commit or workflow-file changes, especially modifications to CI configuration files. Teams should also review CI secrets for exposure, rotate any credentials that may have transited affected pipelines, and apply branch-protection rules requiring review before changes to workflow definitions can merge.
Defenders should prioritize detection over the coming days by auditing recent commits and GitHub Actions workflow changes for unfamiliar scripts or outbound network calls, and by rotating secrets used in any pipeline that may have executed attacker-modified code.
Source: The Hacker News
Read the original report ↗This is an original summary written by CyberSkii for reference. All rights to the underlying reporting remain with the publisher linked above.