WorkOS data breach linked to ARTEX AI offensive framework
Hackers used an automated AI penetration testing suite to access internal sales records, though no production systems were compromised.
WorkOS, a developer infrastructure company, confirmed a security incident where threat actors accessed business contact details and customer communication summaries. The attacker utilized an artificial intelligence-powered offensive security framework to automate reconnaissance and exploitation attempts against the company's internal tools. While sensitive data was exposed, WorkOS stated that no app data or production services were accessed or affected during the breach.
Internal Sales App Compromised
The unauthorized access targeted a specific application used by WorkOS sales teams rather than public-facing products. According to an incident alert released on October 30, 2026, the following categories of information were retrieved:
- Business contact details including names, email addresses, job titles, and company affiliations from CRM systems
- Customer account histories covering product usage and connection or directory health status
- Customer communications comprising summaries and excerpts from sales calls, support discussions, and Slack conversations
- Business account and relationship information detailing account histories and customer relationship dynamics
Following the detection of the intrusion, WorkOS immediately disabled external access to the affected application and reviewed its internal access controls. The company contracted an independent forensics and incident response firm to track the unauthorized activity and determine the full scope of the impact. WorkOS promised to notify affected customers directly and warned users to remain vigilant for phishing or social engineering attempts leveraging the exposed information.
ARTEX AI Likely Powering the Attack
WorkOS founder Michael Grinich identified the specific tool used in the breach as the ARTEX AI penetration testing suite. Released earlier this year by Chinese developers, the platform was initially available as an open-source project for offensive security before becoming private after it was used to target South Korean banks.
Despite being privatized, several forks of the ARTEX code remain available online. The framework has been observed in multiple incidents recently, including a global campaign targeting 200 organizations by the cyber-espionage group "Vapor Panda" and attacks on the cybersecurity non-profit DIVD last September. Grinich noted that investigators identified specific user agents such as 'artex-enrich/1.0' and 'norma/0.4' in web requests originating from the threat actor.
Grinich assured stakeholders that the cybersecurity incident has now been fully contained. He promised to share additional information regarding the attack once the investigation is complete.
