MLFLOW · CVE-2026-0596 · AI agents & MCP
FIX → 3.9.0
When MLflow serves a model with enable_mlserver=True, the model_uri is placed into a bash -c shell command without sanitization. A model URI containing shell substitutions such as $() or backticks lets an attacker run commands of their choosing.
first public Mar 31, 2026 · 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 that serve MLflow models with enable_mlserver=True, where the model_uri can be influenced by a user or comes from a directory that lower-privileged users can write to. A higher-privileged serving service can be escalated through such a directory. Teams on 3.9.0 or later, or not using enable_mlserver, are not affected.
Not affected if fixed-version, mlserver-disabled, trusted-model-uri
Who cares ml-platform, mlflow, model-serving, python
pip show mlflow and compare the version against 3.9.0.enable_mlserver=True or enable_mlserver.Upgrade to 3.9.0
A command injection vulnerability exists in mlflow/mlflow when serving a model with enable_mlserver=True. The model_uri is embedded directly into a shell command executed via bash -c without proper sanitization. If the model_uri contains shell metacharacters, such as $() or backticks, it allows for command substitution and execution of attacker-controlled commands. This vulnerability affects the latest version of mlflow/mlflow and can lead to privilege escalation if a higher-privileged service serves models from a directory writable by lower-privileged users.
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.
mlflow 3.9.0GitHub Advisory Database, CC-BY 4.0 · NVD data is provided by NIST and is public domain. Posts quoted belong to their authors.