What happened in the Kernel security incident?
In its public notice, Kernel says an attacker used a compromised Kernel API key on October 24, 2026 to probe its systems and read internal profiles. On October 27 the attacker gained a staff login to Kernel's logging provider and downloaded data that included a small number of Kernel API keys and third-party secrets from app deployments.
Kernel detected the incident on November 2. On November 3 the attacker used a separately compromised hosting-provider credential to reach four GPU servers, three running Kernel browser sessions and one development server with no customer data. Kernel says the breach affected 0.28% of its customers, assesses the actor as financially motivated and running a cryptomining campaign, and says the affected systems were contained, eradicated and recovered.
Who is affected by the Kernel incident?
- 0.28% of Kernel customers, whom Kernel notified directly.
- Kernel API keys and third-party secrets from app deployments that were captured in the logging provider.
- Four GPU servers reached on November 3, one of which held no customer data.
How to check if you're affected
- Look for Kernel's direct notification to your organization.
- Request the attacker's technical fingerprint from security@kernel.sh to search your own logs.
- Check whether your app deployments wrote secrets into logs that a logging provider stores.
What to do now
- Follow the remediation steps in Kernel's notification.
- Rotate any third-party secrets your deployments exposed, at each issuer.
- Use scoped API keys, as Kernel recommends, so one key can't reach everything.
Why secrets in logs are a breach path
The Kernel chain ran through a logging provider: once an attacker had a staff login there, any secret an application had written to its logs was readable. Logs are rarely treated as a secret store, so they're rarely encrypted, scoped or rotated like one. Scrub credentials before they're logged, keep log retention short for high-risk services, and treat a logging provider's access as access to whatever your apps print. Secret scanning should cover log and config output, not just source code.
Where Nullify fits
How Nullify Covers This Class of Attack
Nullify finds credentials in code and verifies which ones still work with the provider that issued them (AWS, GCP, Azure, GitHub, GitLab, Stripe and Slack), so live keys rank first, then routes rotation to each owner. Rotation happens with the provider; Nullify doesn't rotate keys.
These secrets were stored in the platform, not in your code. Nullify's live-key check helps where the same keys also sit in your repositories or config.
Frequently Asked Questions
What happened in the Kernel security incident?
An attacker used a compromised Kernel API key in late October 2026, then a staff login at Kernel's logging provider to download some customer API keys and deployment secrets, and later a hosting credential to reach GPU servers. Kernel says 0.28% of customers were affected.
What did the attacker want in the Kernel breach?
Kernel assesses the actor as financially motivated and running a cryptomining campaign, which fits the November 3 access to four GPU servers. The downloaded keys and secrets could also be reused against other services, so affected customers should rotate them.
How can teams avoid leaking secrets through logs?
Scrub credentials before they reach logs, scope API keys to one job, keep retention short for sensitive services, and scan log and config output for secrets as well as source code. Kernel's chain shows how a logging provider becomes a secrets store.
Update Log
- Nov 17, 2026: page published from Kernel's November 4 public notice.