Path Traversal Vulnerability in MLflow by Databricks
CVE-2026-53995

8.8HIGH

Key Information:

Vendor

Mlflow

Status
Mlflow
Vendor
CVE Published:
19 October 2026

Badges

👾 Exploit Exists🟡 Public PoC

What is CVE-2026-53995?

MLflow versions 3.17.1 and earlier are susceptible to a path traversal vulnerability within the FileStore tracking backend. This issue allows unauthenticated attackers to manipulate filesystem paths by sending specially crafted trace IDs containing traversal sequences to the POST /api/3.0/mlflow/traces endpoint. By exploiting the FileStore.start_trace() method, attackers can write files such as trace_info.yaml, request_metadata, and tag files with malicious data, potentially outside the intended experiment trace directory. This vulnerability is particularly concerning for systems using FileStore as their default backend before version 3.7.0. In later versions, if FileStore is not specifically enabled, the risk is mitigated as the default changes to a SQLite database.

Affected Version(s)

mlflow 0 <= 3.17.1

Exploit Proof of Concept (PoC)

PoC code is written by security researchers to demonstrate the vulnerability can be exploited. PoC code is also a key component for weaponization which could lead to ransomware.

CVE-2026-53995-mlflow-path-traversal
Created by SaiTeja-Erukude

References

https://github.com/SaiTeja-Erukude/CVE-2026-53995-mlflow-...
technical-descriptionexploit
https://github.com/mlflow/mlflow/pull/26842
issue-tracking
https://www.vulncheck.com/advisories/mlflow-path-traversa...
third-party-advisory

CVSS V4

Score:
8.8
Severity:
HIGH
Confidentiality:
None
Integrity:
High
Availability:
Low
Attack Vector:
Network
Attack Complexity:
Low
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
None

Timeline

  • 🟡

    Public PoC available

  • 👾

    Exploit known to exist

  • Vulnerability published

  • Vulnerability Reserved

Credit

Sai Teja Erukude
.