Multiple people who appear to be employees of Microsoft have exposed sensitive login credentials to the company’s own infrastructure on GitHub, potentially offering attackers a gateway into internal Microsoft systems, according to a cybersecurity research firm that found the exposed credentials.

Microsoft confirmed the data exposure when contacted by Motherboard. Microsoft owns GitHub.

“We continue to see that accidental source code and credential leakages are part of the attack surface of a company, and it’s becoming more and more difficult to identify in a timely and accurate manner. This is a very challenging issue for most companies these days,” Mossab Hussein, chief security officer at cybersecurity firm spiderSilk which discovered the issue, told Motherboard in an online chat. SpiderSilk has previously discovered an exposed list of Slack channels belonging to Electronic Arts; the personal information of WeWork customers uploaded by WeWork developers; and that education giant Elsevier exposed users’ passwords.

Do you know about any other sensitive data exposures? We’d love to hear from you. Using a non-work phone or computer, you can contact Joseph Cox securely on Signal on +44 20 8133 5190, Wickr on josephcox, or email joseph.cox@vice.com.

Hussein provided Motherboard with seven examples in total of exposed Microsoft logins. All of these were credentials for Azure servers. Azure is Microsoft’s cloud computer service and is similar to Amazon Web Services. All of the exposed credentials were associated with an official Microsoft tenant ID. A tenant ID is a unique identifier linked to a particular set of Azure users. One of the GitHub users also listed Microsoft on their profile.

Three of the seven login credentials were still active when spiderSilk discovered them, with one seemingly uploaded just days ago at the time of writing. The other four sets of credentials were no longer active but still highlighted the risk of workers accidentally uploading keys for internal systems.

Sign up for Motherboard’s daily newsletter for a regular dose of our original reporting, plus behind-the-scenes content about our biggest stories.

Microsoft refused to elaborate on what systems the credentials were protecting when asked multiple times by Motherboard. But generally speaking, an attacker may have an opportunity to move onto other points of interest after gaining initial access to an internal system. 

One of the GitHub profiles with exposed and active credentials makes a reference to the Azure DevOps code repository. Highlighting the risk that such credentials may pose, in an apparently unrelated hack in March attackers gained access to an Azure DevOps account and then published a large amount of Microsoft source code, including for Bing and Microsoft’s Cortana assistant.

Regarding the fresh credential exposure, a Microsoft spokesperson told Motherboard in an email that “We’ve investigated and have taken action to secure these credentials. While they were inadvertently made public, we haven’t seen any evidence that sensitive data was accessed or the credentials were used improperly. We’re continuing to investigate and will continue to take necessary steps to further prevent inadvertent sharing of credentials.”

Subscribe to our cybersecurity podcast, CYBER. Subscribe to our new Twitch channel.