An attacker used a compromised GitHub account and leaked credentials for four weeks, copying customer contacts, contract terms, SQL editor text and encrypted API keys. PingCAP says the databases themselves were not reached.
Northflank tells every customer to rotate secrets after a breach
EditorialBy TrustList Editorial
The cloud platform says to assume all environment variables were decrypted. Customers have since reported their AI provider keys being used, and audit logs are now on for every account, free users included.
- Cloud Security
- Cybersecurity
- DevOps
- PaaS
- +1 more
About Northflank tells every customer to rotate secrets after a breach
Northflank tells every customer to rotate secrets after a breach
30 October 2026: Northflank, a platform that runs customers' containers, databases, jobs and builds, has told all of its users to assume their environment variables have been decrypted and to rotate every secret and credential they store with it. Within hours, users reported that AI API keys held on the platform were being used without their consent, and Northflank's status page still showed major outages across networking, certificates, services, addons, jobs, builds and logs on the morning of 31 October.
Not yet independently verified. The timeline is Northflank's own, from its status page; it has not yet said how the attacker got in or how many customers were affected, and its root cause analysis is still to come. We will update this when it can be confirmed, and remove this note.
Update, 13 November 2026: Northflank's status page added updates on 2 and 6 November 2026. It says secrets were first known to be decrypted between 02:39 and 03:08 UTC on 30 October, that the first alert came at 11:43 UTC, and that the flaw was patched within 60 minutes of notification. It reports no evidence that workloads or workload data were accessed, and an independent forensic review is under way.
Eighty minutes from alert to "assume decrypted"
Northflank's own timeline starts with a threat alert from the security firm Wiz at 11:43 UTC on 30 October. Engineers reviewed it a minute later, deployed a first network policy patch at 12:02 and further patches at 12:50 and 13:05, and asked Wiz's incident response team for help at 13:00. At 13:40 the status page told users to rotate environment variables immediately, and at 13:58 it went further: "Assume all environment variables have been decrypted. Disable/rotate all keys."
At 15:12 Northflank enabled secret rotation for Postgres and MongoDB addons for all users and listed the three places secrets live on the platform: secrets set on workloads, secret groups and global secrets. By then, it said, multiple users were confirming that their AI keys had been used. A 19:28 update said the investigation had no further confirmed findings and the guidance was unchanged.
AI keys used, and a link to a wider campaign
At 23:38 Northflank reported unauthorised use of AI API keys supplied to the platform and said the reports "may be consistent with the broader pattern of attacks targeting AI credentials", pointing to IP addresses listed in Wiz's "Black October" report. At 01:04 on 31 October it switched on audit logs for every customer, free accounts included. Northflank has not said how the attacker reached the secrets, whether data in databases was read, or how many accounts are affected.
The rotation list for teams on Northflank
Northflank's guidance amounts to a full rotation rather than a selective one:
- revoke each credential with the provider that issued it, generate a replacement, and update the secret and every workload that uses it;
- cover workload secrets, secret groups and global secrets, and rotate database credentials with the new Postgres and MongoDB rotation;
- disable public access on any addon where it was enabled;
- check AI provider usage and billing logs for activity since 30 October, and report anything unexpected to Northflank and to the provider.
Keys that were rotated before the AI misuse reports may still have been used in the hours before; billing logs show when. The audit logs now enabled on every account are the first place to look for actions taken inside the platform itself.
Related on TrustList:
- Wiz ties break-ins at 14 AI developer platforms to stolen model keys
- Heroku ACM switches new certificates from RSA to ECDSA P-256 on 21 December
- Heroku-22 reaches end of life on 30 April 2027; move apps now
Sources
Categories & features
- Cloud Security
- Cybersecurity
- DevOps
- PaaS
- Data Security
TrustList Weekly
The week in software and IT, in one email
The news that matters to buyers, new rankings and our own research. Every Thursday, free, and easy to leave.
You may be interested in
The hosting platform paused all deployments after finding access to stored environment variables, some decrypted. It links the attack to a campaign against AI inference providers and tells every user to roll keys.
A hijacked contributor account replaced the macOS, Linux and Windows binaries of five releases on GitHub; v0.16.0 is a clean, attested rebuild, and anyone who installed in the window should rotate secrets.
Heroku repeated the deadline alongside a 20 October update to all three stacks; buildpack apps get fixes automatically, but apps on the container stack must rebuild their Docker images.
The Thoma Bravo-owned email security firm adds Varonis's data security platform. Varonis tells customers nothing changes before the deal closes, expected in the first half of 2027.
CVE-2026-119115 let users limited to a namespace issuer get certificates, even subordinate CAs, from a cluster issuer; v1.10.1 fixes it but stops renewals that lack an exact issuerRef.kind.
Wiz Research tracked more than 13,500 planted ci-sync branches from over 850 compromised accounts; GitHub's approval hold stops the workflows in public repositories but not in private ones or GitHub Enterprise Server.
Four AWS bulletins on 22 October: Strands Agents A2A servers leaked conversation history between callers, SAM CLI's Terraform hook allowed command injection, Athena exposed query text, and aws-c-mqtt could crash clients.
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- AI ModelsBenchmark scores and community ratings for every major model.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.