GHSA-r23q-823p-vmf7 BIT-mlflow-2025-15379 CVE-2025-15379 PYSEC-2026-423
MLflow Command Injection vulnerability
- Severity
- 10.0 / 10
- Critical · CVSS 3.0
- Exploit chance (EPSS)
- 2.40%
- Chance of exploitation in the next 30 days. Higher than 84.04% of scored vulnerabilities.
- Known exploited
- No
- Not in CISA's catalog
- Weakness
- CWE-77 Improper Neutralization of Special Elements used in a Command ('Command Injection')
- Fix
3.8.1- Upgrade mlflow to this version or later
Description
A command injection vulnerability exists in MLflow's model serving container initialization code, specifically in the _install_model_dependencies_to_env() function. When deploying a model with env_manager=LOCAL, MLflow reads dependency specifications from the model artifact's python_env.yaml file and directly interpolates them into a shell command without sanitization. This allows an attacker to supply a malicious model artifact and achieve arbitrary command execution on systems that deploy the model. The vulnerability affects versions 3.8.0 and is fixed in version 3.8.1.
Affected versions
PyPI
| Package | Vulnerable | Fixed in |
|---|---|---|
| mlflow | < 3.8.1 |
3.8.1 |
How to fix
Move to the fixed version or later with your package manager, update the lockfile, and run your tests. If the package only arrives through another dependency, upgrade that dependency first. If none of its releases pulls a fixed version yet, force the version with your package manager's override.
mlflow PyPI → 3.8.1
- pip
pip install --upgrade "mlflow>=3.8.1"- poetry
poetry add "mlflow>=3.8.1"- uv
uv add "mlflow>=3.8.1"
Severity breakdown
CVSS 3.0 10.0
CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
- Attack vector
- Network
- Attack complexity
- Low
- Privileges required
- None
- User interaction
- None
- Scope
- Changed
- Confidentiality
- High
- Integrity
- High
- Availability
- High
Weakness
CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection')
The product constructs all or part of a command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended command when it is sent to a downstream component.
How to prevent it
- Architecture and Design: If at all possible, use library calls rather than external processes to recreate the desired functionality.
- Implementation: If possible, ensure that all external commands called from the program are statically created.
- Implementation: Assume all input is malicious. Use an "accept known good" input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does. When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, "boat" may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as "red" or "blue." Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code's environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright.
- Operation: Run time: Run time policy enforcement may be used in an allowlist fashion to prevent use of any non-sanctioned commands.
- System Configuration: Assign permissions that prevent the user from accessing/opening privileged files.
References
Advisories
Packages
Other links
Timeline
- Published
- Last updated by the source
- Latest EPSS score
Related advisories
| Advisory | Severity | Why it's related |
|---|---|---|
| GHSA-7gwp-5pfp-969j MLflow: Unauthenticated full-read SSRF in webhook delivery: _validate_webhook_url bypassed via unvalidated HTTP redirects (and DNS rebinding) | Critical Known exploited | Also affects mlflow |
| GHSA-26p8-2jq9-3vq9 MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules | Critical | Also affects mlflow |
| GHSA-j7qq-vr8c-xvw3 MLflow CrewAI autologging records LLM provider API keys in logged trace spans | Medium | Also affects mlflow |
| GHSA-m65p-9j7q-6w33 MLflow vulnerable to arbitrary file read on unauthenticated servers via prompt model version source | High | Also affects mlflow |
| GHSA-2wqp-q3c6-ccj5 MLflow: SSRF on unauthenticated MLflow servers via third-party scorer kwargs in the scorer invoke, register, and GenAI evaluate endpoints | Medium | Also affects mlflow |
| GHSA-wqmh-qccw-pqwf MLflow FileStore trace IDs allow path traversal and arbitrary file writes | High | Also affects mlflow |
| GHSA-gqvg-gmmx-x4hm MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False safety control bypassed by mlflow.statsmodels flavor — RCE via crafted model artifact | High | Also affects mlflow |
| GHSA-h7x2-h6g9-p789 MLflow AI Gateway permits SSRF through an unvalidated api_base | High | Also affects mlflow |
MergeHarbor checks every repository's dependencies against this database, ranks what matters by exploitability and reachability, routes it to the owning team and opens the fix pull request.
Check your repos with MergeHarbor