Early warning
Researchers say attackers use secrets in Docker Hub images within eight seconds
Stony Brook University researchers published 300 honeypot images on eight public registries and measured planted AWS, OpenAI and Anthropic keys being used within seconds, with over 140,000 AWS API calls from 5,000 IP addresses.
Written by a model from the sources below. Verify against them.
- Kind
- Pending CVE
- Vendor
- Docker
- Product
- Docker Hub
- Threat actor
- ContainerBots
- Status
- Corroborated
- First seen
Summary
Stony Brook University researchers published 300 honeypot images on eight public registries and measured planted AWS, OpenAI and Anthropic keys being used within seconds, with over 140,000 AWS API calls from 5,000 IP addresses. The researchers said they reported the issue to Docker in May and received no follow-up beyond an acknowledgment.
What to do
Rotate any cloud, AI API and database credentials that were ever stored in published Docker Hub images.
Scan images for embedded secrets before pushing them to any public registry.
Remove secrets from image layers and rebuild images without them.
Timeline
- Made public by BleepingComputer
- First seen by cve.now
Evidence 1
- PressBleepingComputer
“Credentials and access secrets exposed in public Docker Hub images are being extracted and exploited by attackers within as little as eight seconds of publication.”
+1 more: BleepingComputer on Bluesky
Corrections
None so far.
Something wrong on this page? Report an error. Every fix is listed here.
Questions about SZB-2cac7202a9
Has Docker confirmed it?
BleepingComputer reports it. No statement from Docker has reached cve.now yet.Source: BleepingComputer
What should Docker and Docker Hub users do?
Rotate any cloud, AI API and database credentials that were ever stored in published Docker Hub images. Scan images for embedded secrets before pushing them to any public registry. Remove secrets from image layers and rebuild images without them.