MLFLOW · CVE-2023-6975 · AI agents & MCP
FIX → 2.9.2
MLflow, an open-source platform for managing machine learning experiments and models, has a path traversal flaw in versions before 2.9.2. An attacker can use it to run commands on the vulnerable machine and read data and model information.
first public Dec 20, 2023 · 1 public exploit · 2 sources
Check it with your coding agent
It reads this page and tells you if your code or app is affected.
Teams running an MLflow tracking server or MLflow-based model registry on a machine reachable by others are exposed. Local notebook use on a single private machine carries less risk. Teams on managed MLflow services with the vendor patching the platform can skip the upgrade but should confirm with the vendor.
Not affected if local-only-use, managed-hosting, version-2.9.2-or-later
Who cares ml-platform, mlflow, python, self-hosted
pip show mlflowmlflow in your requirements.txt, pyproject.toml or lock files and see whether the pinned version is below 2.9.2A malicious user could use this issue to get command execution on the vulnerable machine and get access to data & models information.
This page as .md · .json · CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
GitHub Advisory Database, CC-BY 4.0 · . Posts quoted belong to their authors.
Are you affected?
Ask from here. InfraStack checks the affected versions against your stack and answers in your inbox.
InfraStack checks the affected versions against your stack and answers in Slack, with the evidence and the fix.
InfraStack checks it against your stack and answers on your report, with the evidence and the fix.
Upgrade to 2.9.2
mlflow to 2.9.2 or later