- VERCEL-70Negative
GoPlus Security 🚦 (@GoPlusSecurity)
🚨 A Vercel Employee Was Hit by an AI Supply Chain Attack — Check Your Environment Variables’ Sensitive Flags to Prevent Data Exfiltration I. Attack Chain Analysis 1. Initial supply chain breach: The attacker compromised a third-party AI tool, Context[.]ai, and used it as a pivot to launch a supply chain attack. 2. OAuth credential hijacking: A Vercel employee had previously authorized this AI tool via Google Workspace. The attacker leveraged this to steal the associated OAuth access token. 3. Account takeover by bypassing MFA: Using the legitimately stolen OAuth token, the attacker bypassed enterprise SSO and MFA, successfully taking over the employee’s Google Workspace account. 4. Internal lateral movement: With valid internal identity, the attacker rapidly moved laterally across Vercel’s infrastructure, reaching production environments and business systems. 5. Attack execution: The attacker performed system enumeration and exfiltrated core data at scale. II. Remediation & Defense Guide 1. Credential rotation (critical): Assume all credentials stored in non-Sensitive environment variables are compromised. Immediately rotate high-privilege secrets such as GitHub/NPM tokens, database passwords, and cloud AK/SK. 2. Enforce Sensitive flags: Audit all project configurations and ensure the Sensitive flag is strictly enabled for all confidential environment variables. 3. Log audit & access cleanup: Review recent Vercel logs for API calls and IP activity, and revoke unnecessary or stale third-party integrations and OAuth grants. Incident report below 👇 https://vercel.com/kb/bulletin/vercel-april-2026-security-incident