GitHub Breach, npm Supply Chain Attack, and Mass Data Leaks Dominate May 20

Events tracked
157
Critical exposure
53

Summary

Today's threat landscape is defined by a convergence of supply chain and credential theft incidents, with a confirmed breach at GitHub via a poisoned VS Code extension and a sophisticated npm package compromise targeting CI/CD secrets. These events underscore a shift in attacker focus toward developer toolchains as primary vectors. Simultaneously, a high volume of alleged data leaks and breaches -- 53 critical exposure events -- continues to pressure government and logistics sectors globally, with the United States, France, and India as primary targets.

Today's developments

GitHub confirms internal repository exfiltration. GitHub disclosed that an employee device was compromised through a malicious Visual Studio Code extension, leading to the theft of internal repositories. Multiple threat actors, including groups claiming to be "Lazarus" and "TeamPCP," have since alleged possession of GitHub source code and records. Industry researchers note that this incident highlights the growing risk of third-party developer tools as attack surfaces, especially when extensions are sourced from unverified channels.

Microsoft disrupts malware-signing service and warns of npm supply chain attack. Microsoft announced the takedown of a malware-signing-as-a-service operation attributed to a threat actor it tracks as Fox Tempest, which abused Microsoft's Artifact Signing system to sign ransomware payloads. Separately, Microsoft's security team detailed the "Mini Shai Hulud" campaign, where compromised @antv npm packages deployed a payload that steals CI/CD credentials from Linux environments during npm install. The malware targets secrets across GitHub, AWS, Kubernetes, Vault, npm, and 1Password platforms, representing a direct threat to software development pipelines.

Mass data exposure events target logistics, government, and financial sectors. Today's 53 critical data exposure events include alleged breaches of major logistics firms: United Parcel Service (UPS) and iShip in the US, and Kintetsu World Express in Singapore. Government entities in Uruguay, the Philippines, Kuwait, Algeria, Georgia, and Indonesia are also allegedly compromised. In the financial sector, an alleged breach of Bank of India and a claim involving 10.6 million Indian stock market investors were reported. Multiple threat actors -- including huntertrace, The BlackH4t MD-Ghost, and S10 -- are responsible for a wave of alleged leaks targeting Israeli and French citizen data, as well as US and UK driving license records.

7-Eleven confirms breach after ShinyHunters claims. In a confirmed incident, 7-Eleven notified authorities of a breach discovered on April 8, where attackers accessed systems storing franchisee documents. This aligns with earlier claims by the ShinyHunters group and reinforces the persistent threat to retail and franchise operations.

Threat landscape signals

Actor concentration and targeting patterns. The HellR00ters Team remains the most prolific actor today with 42 events, primarily defacements, but the most impactful incidents are driven by smaller, specialized groups. Huntertrace alone is responsible for five alleged data breaches targeting logistics and telecom firms across the US, Italy, Benin, and Costa Rica. The BlackH4t MD-Ghost continues to focus on Israeli and government targets, while S10 is running a multi-country data leak campaign against France, Israel, and the Philippines.

Supply chain and CI/CD attacks are the new frontier. The GitHub and npm incidents, combined with Microsoft's disclosure of the Fox Tempest signing service, signal that attackers are systematically targeting the software supply chain. Defenders should prioritize auditing third-party extensions, enforcing strict npm package vetting, and implementing just-in-time credential models for CI/CD pipelines. The introduction of Microsoft's open-source RAMPART and Clarity tools for AI agent security testing reflects the industry's recognition that agentic AI introduces new attack surfaces that require dedicated tooling.

Geographic and sectoral pressure points. The United States remains the most targeted country with 36 events, followed by France (12) and India (9). Government administration and transportation/logistics are the most affected sectors, with education and financial services also under sustained pressure. The high volume of alleged sales of identity documents (driving licenses, ID cards) from the US, UK, Germany, and France suggests a mature underground market for personally identifiable information that will likely fuel future fraud and phishing campaigns.

All incidents are reported as alleged claims by threat actors and have not been independently verified by GrayscaleInsight.

Threat intelligence is reported for security awareness purposes only and does not constitute endorsement of any actor, group, or activity.

Recent editions