Introduction
Supply chain attacks on Docker Hub are becoming increasingly sophisticated. In 2026, two major incidents—first involving Trivy, then Checkmarx KICS—demonstrated how stolen publisher credentials can lead to malicious container images being pushed through legitimate publishing workflows. In both cases, attacker-controlled code was added to official repositories, exfiltrating sensitive data such as credentials and cloud resource names. This guide shows you how to systematically respond to such an attack, using the KICS incident as a practical example. You’ll learn how to identify compromised images, assess the damage, rotate exposed secrets, purge malicious artifacts, and implement long-term protections like digest pinning. By following these steps, you can minimize the impact of future supply chain compromises and harden your CI/CD pipeline against similar threats.
What You Need
- Access to Docker Hub or your image registry (e.g., Docker Hub account, registry credentials)
- CI/CD logs and history (e.g., Jenkins, GitLab CI, GitHub Actions job logs)
- A list of all Docker images pulled from
checkmarx/kics(or affected repository) in your environment - Tools to inspect image digests (e.g., Docker CLI, Skopeo, or a registry API client)
- Access to credential stores (e.g., password managers, secret vaults, AWS Secrets Manager)
- A secure communication channel to notify your team and affected stakeholders
Step-by-Step Response Guide
Step 1: Identify Compromised Images by Digest
The first step is to determine which tags and digests were affected during the attack window. In the KICS incident, the threat actor overwrote five existing tags (latest, v2.1.20, v2.1.20-debian, alpine, debian) and created two new tags (v2.1.21, v2.1.21-debian). Check your Docker pull history or CI logs for any of these tags. Use the following malicious digest list from the official post to verify:
- Alpine, v2.1.20, v2.1.21 →
sha256:2588a44890263a8185bd5d9fadb6bc9220b60245dbcbc4da35e1b62a6f8c230d(index manifest) - Debian, v2.1.20-debian, v2.1.21-debian →
sha256:222e6bfed0f3bb1937bf5e719a2342871ccd683ff1c0cb967c8e31ea58beaf7b - Latest →
sha256:a0d9366f6f0166dcbf92fcdc98e1a03d2e6210e8d7e8573f74d50849130651a0
For each image you pulled, run docker inspect <image> and compare the RepoDigests field against the above list. If any match, you have a compromised image.
Step 2: Assess the Exposure Window
The attack began around 12:35 UTC on April 22, 2026. Determine the exact times your CI pipelines ran KICS scans using the affected tags. Any scan executed after that timestamp may have exfiltrated data. The malicious binary collected scan output—including secrets, credentials, and cloud resource names—and encrypted it, then sent it to audit.checkmarx[.]cx with the User-Agent KICS-Telemetry/2.0. Review your network logs for outbound connections to that domain. Also check for any unusual DNS queries or traffic patterns in your cloud environment.
Step 3: Rotate Exposed Credentials Immediately
If your CI ran KICS against any repository containing credentials, cloud provider keys, or API tokens during the exposure window, assume those credentials are compromised. Rotate them now. This includes:
- Database passwords and connection strings
- Cloud service provider access keys (AWS, Azure, GCP)
- Third-party API keys and secrets
- SSH keys and certificates
Use a structured process: first invalidate the old credential, then generate a new one, and finally update any services that depend on it. Consider automating credential rotation with a secrets management tool.
Step 4: Purge Malicious Images from All Locations
Remove the compromised images from everywhere they might be stored:
- Local Docker cache: Run
docker rmi <image-digest>to delete the image layer. - CI runners: Connect to each runner and clear the Docker cache or reset the host.
- Pull-through registries: If you use a proxy cache (e.g., Docker Hub mirror), delete the cached tags and digests.
- Artifact registries: Remove any copies stored in ECR, GCR, or private registries.
After purging, verify by pulling the image again (using a known clean digest) and checking that no malicious files remain.
Step 5: Pin Your CI to Image Digests, Not Tags
Tags are mutable and can be overwritten. To prevent future supply chain attacks, modify your CI/CD pipelines to use image digests instead of tags. For example, change image: checkmarx/kics:latest to image: checkmarx/kics@sha256:<known-good-digest>. Use the digest that was published before the incident. This ensures your builds always use the exact same image content, even if an attacker pushes a new malicious version to the same tag. Document the method in your team’s build guidelines.
Step 6: Implement Monitoring and Alerting for Supply Chain Risks
Set up automated checks to detect anomalies in your container image supply chain. Consider:
- Image signature verification: Use tools like Docker Content Trust or Notary to verify content integrity.
- Digest change detection: Monitor for changes to image digests in your CI logs and alert on unexpected updates.
- Network telemetry: Log outbound connections from CI runners and flag new or unexpected destinations (like
audit.checkmarx[.]cx). - Registry audit logs: Regularly review Docker Hub audit logs for unusual pushes or credential usage.
Integrate these alerts into your incident response platform for rapid action.
Conclusion and Tips
Supply chain attacks are evolving, but a systematic response can contain damage and prevent recurrence. The KICS and Trivy incidents highlight how stolen publisher credentials can be weaponized with minimal infrastructure compromise. Here are some closing tips:
- Always verify authenticity. Use signed images and cryptographic verification whenever possible.
- Trust digests, not tags. Pinning to digests is your strongest defense against tag overwriting.
- Rotate credentials proactively. Implement periodic rotation of CI secrets and use short-lived tokens.
- Educate your team. Share these incidents and the response steps with all developers and operators.
- Practice incident drills. Conduct tabletop exercises to simulate a supply chain attack response.
By adopting these measures, you reduce the attack surface and increase resilience against future compromises. Share this guide with your security team and update your incident response playbook accordingly.