Aikido says hackers used 4,837 compromised GitHub accounts to probe the write permissions of 60,000 repositories
駭客利用逾4,800個遭入侵GitHub帳號,鎖定6萬個儲存庫竊取雲端憑證
Security vendor Aikido Security disclosed a large-scale GitHub credential theft campaign codenamed wfx-loot: since November 10, hackers have used 4,837 compromised accounts to probe the write permissions of 61,764 repositories, and at least 1,835 repositories have executed malicious programs. The attack does not exploit a vulnerability in GitHub Actions itself; rather, it impersonates these accounts to modify workflows or implant malicious shell scripts, stealing GitHub tokens, AWS, Google Cloud, and Azure cloud credentials, and API keys for AI services such as OpenAI and Anthropic. It can also create long-lived AWS access keys, modify IAM role trust policies, and add Azure federated identity credentials, potentially retaining access to victims' cloud resources after the original GitHub tokens are revoked.