CVE-2026-5817
Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends
CVSS Score
8.2
EPSS Score
0.0%
EPSS Percentile
3th
The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trust_remote_code=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.from_pretrained() to import and execute arbitrary Python files included in any model pulled from an OCI registry, resulting in arbitrary code execution on the Docker host as the Docker Desktop user when inference is triggered. Any container on the Docker network can trigger this by calling the model-runner.docker.internal API to pull a malicious model and request inference.
| CWE | CWE-829 |
| Vendor | docker |
| Product | docker desktop |
| Ecosystems | |
| Industries | Technology |
| Published | May 22, 2026 |
| Last Updated | May 27, 2026 |
Stay Ahead of the Next One
Get instant alerts for docker docker desktop
Be the first to know when new high vulnerabilities affecting docker docker desktop are published — delivered to Slack, Telegram or Discord.
Get Free Alerts →
Free · No credit card · 60 sec setup
CVSS v3 Breakdown
CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H Attack Vector
Local
Attack Complexity
Low
Privileges Required
Low
User Interaction
Required
Scope
Changed
Confidentiality
High
Integrity
High
Availability
High
Affected Versions
Docker / Docker Desktop
4.62.0 < 4.68.0
References
Credits
David Rochester (@davidrxchester) Nicholas Gould (@gouldnicholas)