CVE-2026-127202 MLflow < 3.18.0 Path Traversal via model_id in log_outputs API
Unauthenticated attackers can write files and directories anywhere on MLflow servers using the FileStore backend via a crafted model_id.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Harmful if only installed
- No
- Needs a specific call
- No
- Fix
- None listed
What is CVE-2026-127202?
CVE-2026-127202 is a medium-severity path traversal vulnerability in mlflow, published on 12 Nov 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Official description
MLflow before 3.18.0 using the FileStore backend contains a path traversal vulnerability that allows unauthenticated remote attackers to write files outside the intended run directory by supplying a malicious model_id parameter to the log outputs API. Attackers can exploit the unvalidated model_id containing ../ sequences via the POST /api/3.0/mlflow/runs/outputs endpoint to create directories and write a meta.yaml file at arbitrary filesystem locations, limited only by the server account's permissions. FileStore was the default backend before 3.7.0 and is opt-in afterward through MLFLOW_ALLOW_FILE_STORE.
Weakness: CWE-22
Which versions are affected by CVE-2026-127202?
1 open source package ship an affected version, including mlflow.
An affected version is exposed as installed. No special setting or call is needed.
| Package | Affected | Fixed in |
|---|---|---|
mlflow PyPIpkg:pypi/mlflow | All versions | 3.18.0 |
How do I fix CVE-2026-127202?
Upgrade mlflow to 3.18.0 or later.
Upgrade, and resolve every user-supplied path against an allowed base folder before opening it. More about Path traversal.
Is CVE-2026-127202 being exploited?
There is no known exploitation of CVE-2026-127202 in the wild and no public exploit code.
Sources
Also known as GHSA-2265-6gg9-3q46.