PatchSiren cyber security CVE debrief
CVE-2026-116690 mlflow CVE debrief
CVE-2026-116690 debrief: MLflow vulnerable to arbitrary file read on unauthenticated servers via prompt model version source. The MLflow tracking server is vulnerable to arbitrary file read attacks when running without authentication or using basic authentication. An attacker can create a model version tagged as a prompt with a schemeless local path, allowing them to read files from the server's filesystem. This vulnerability allows for potential unauthorized file access and data breaches. Defenders responsible for MLflow server configurations should assess exposure to potential file read attacks and prioritize verification and remediation efforts.
- Vendor
- mlflow
- Product
- Unknown
- CVSS
- HIGH 7.5
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-10-27
- Original CVE updated
- 2026-10-27
- Advisory published
- 2026-10-27
- Advisory updated
- 2026-10-27
Who should care
Defenders responsible for MLflow server configurations, especially those running without authentication or using basic authentication, should assess exposure to potential file read attacks and prioritize verification and remediation efforts.
Why it matters
CVE-2026-116690 is significant for defenders responsible for MLflow server configurations, especially those running without authentication or using basic authentication. The vulnerability allows for arbitrary file read attacks, potentially leading to unauthorized file access and data breaches. Defenders should prioritize verifying MLflow server configurations and updating to version 3.17.0 or later.
- Potential unauthorized file access
- Possible data breaches through file read attacks
- Need for verification of MLflow server configurations
- Priority for updating MLflow to version 3.17.0 or later
Technical summary
The MLflow tracking server is vulnerable to arbitrary file read attacks when running without authentication or using basic authentication. An attacker can create a model version tagged as a prompt with a schemeless local path, allowing them to read files from the server's filesystem. The fix for CVE-2026-2614 in 3.10.0 added a prompt-specific branch to the `CreateModelVersion` handler that rejects `file://` URIs and absolute paths but runs path validation only when the source carries a URL scheme. A schemeless value such as `../../etc`, a plain relative name, `.`, a percent-encoded path, or a Windows drive-letter path is therefore stored unchanged as the model version’s artifact location, and the `model-ersions
Defensive priority
Defenders should prioritize verifying MLflow server configurations, especially those running without authentication or using basic authentication, and assess exposure to potential file read attacks.
Recommended defensive actions
- Verify MLflow server configurations to ensure authentication is enabled
- Assess exposure of MLflow servers to potential file read attacks
- Update MLflow to version 3.17.0 or later
- Review compensating controls for exposed systems while remediation is scheduled and verified
- Check relevant monitoring, detection, and logs for exposed assets that need extra review
- Track exceptions, retest remediated assets, and close the item only after evidence is documented
- Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up
Evidence notes
The CVE record and source item provide details on the vulnerability, including its impact on MLflow tracking servers running without authentication. However, the record of affected versions and remediation efforts is limited.
Sources and references
Verified primary and authoritative sources
-
CVE-2026-116690 CVE Program record
Publisher, destination, and source semantics verified
URL: https://www.cve.org/CVERecord?id=CVE-2026-116690
CVE Program - Official CVE Program record with source-provided CVE metadata.
-
CVE-2026-116690 NVD vulnerability detail
Publisher, destination, and source semantics verified
URL: https://nvd.nist.gov/vuln/detail/CVE-2026-116690
NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.
Supplemental references
-
MLflow vulnerable to arbitrary file read on unauthenticated servers via prompt model version sou
Unverified legacy reference
URL: https://storage.googleapis.com/osv-vulnerabilities/PyPI/GHSA-m65p-9j7q-6w33.json
osv_dev
-
Source reference
Unverified legacy reference
URL: https://github.com/mlflow/mlflow/security/advisories/GHSA-m65p-9j7q-6w33
Supplemental source
-
Source reference
Unverified legacy reference
URL: https://github.com/mlflow/mlflow/pull/25870
Supplemental source
-
Source reference
Unverified legacy reference
URL: https://github.com/mlflow/mlflow/commit/81730a2ce4ebcd60d5d18f68c96f422ad7e30772
Supplemental source
-
Source reference
Unverified legacy reference
URL: https://github.com/mlflow/mlflow
Supplemental source
-
Source reference
Unverified legacy reference
URL: https://github.com/mlflow/mlflow/releases/tag/v3.17.0
Supplemental source
Methodology and review provenance
AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.