Vercel says attackers exploited an OAuth connection tied to Context AI to hijack a corporate Google account, then moved into its internal systems to steal customer app data and API keys.
How attackers gained access
Vercel, the cloud app host used by thousands of developers, said the intrusion began when an employee installed an app from Context AI and linked it to a corporate Google account. Attackers exploited that OAuth connection to take control of the employee's Google account and move into Vercel's internal systems.
Vercel said intruders accessed internal credentials that were not encrypted. The company identified the Context AI connection as the initial vector and has been seeking answers from the software maker about how its app was used to pivot into Vercel’s environment.
A Vercel spokesperson declined to give a total number of affected customers but warned the breach could touch "hundreds of users across many organizations," and said the incident may have downstream effects beyond Vercel itself.
What was taken and what's not affected
Vercel said attackers obtained app data and keys from some customers. Intruders posted a listing in a cybercriminal forum claiming to sell API keys, source code and database data taken from Vercel systems; that listing identified itself with the ShinyHunters name. ShinyHunters later told a cybersecurity news outlet it was not involved in the incident.
Vercel has not publicly linked the forum seller to any specific group and said it has not received a ransom demand or other direct communication from the attacker.
Vercel emphasised that its widely used open-source projects Next.js and Turbopack were not affected. The company said it has contacted customers whose app data and credentials were compromised and urged teams to take immediate steps to protect deployments.
Context AI's role and earlier breach
Context AI confirmed on its website that it experienced a breach in March involving its Context AI Office Suite consumer app. That app provides automation and workflow integrations across multiple third-party services via a separate, unnamed third-party provider.
Context AI said it notified at least one customer following the March breach. After Vercel's disclosure, Context AI began re-evaluating the scope of that earlier incident and whether it links to the access seen inside Vercel's systems.
Supply-chain implications for developers
The Vercel incident is the latest example of a supply-chain attack: compromise one widely used tool and you can harvest credentials and access tokens that grant further entry into other companies' systems. By abusing OAuth connections and integrations, attackers can turn a single third-party app into a bridge between services.
That model matters because developer platforms and automation tools are central to modern software stacks. When a tool that automates workflows or connects multiple cloud services is compromised, credentials and keys that developers rely on can be exposed in bulk. Once an attacker has valid API keys or database credentials, they can access downstream systems without repeatedly breaching each one.
Security teams should review which OAuth apps and third-party integrations have broad privileges and look for service accounts or deployment keys stored in internal systems without encryption. Recommended immediate actions include:
- Rotate exposed API keys and deployment credentials.
- Audit OAuth app permissions and revoke unnecessary integrations.
- Identify and encrypt any service account keys stored in internal systems.
Related Articles
- NSA reportedly using Anthropic's Mythos to scan for vulnerabilities
- Google Workspace Promo Code: Up to 14% Off
- Bank of England to Question Top Lenders on Anthropic’s Mythos Risks
Vercel CEO Guillermo Rauch urged customers to rotate API keys and any credentials marked "non-sensitive," and to follow the company’s guidance for securing deployments.
This article was created with AI assistance.
