Security readout for executives and security teams
Plain-English summary
MLflow can unsafe-extract a tar.gz archive in a way that lets files be written outside the intended cache directory. In shared ML or cluster environments, this could damage workloads, alter sensitive files, or support privilege escalation. The source states versions before v3.7.0 are affected.
Executive priority
Treat this as urgent for shared ML platforms because a crafted artifact could cross tenant or sandbox boundaries. Prioritize internet-accessible, multi-tenant, or privileged MLflow deployments before lower-risk isolated environments.
Technical view
The flaw is in `extract_archive_to_dir` in `mlflow/pyfunc/dbconnect_artifact_cache.py`. Tar member paths are not validated during extraction, enabling path traversal outside the sandbox directory. The CVSS vector is 9.6, network reachable, low complexity, no privileges, user interaction required, and changed scope.
Likely exposure
Organizations running mlflow/mlflow before v3.7.0 are potentially exposed, especially where users or tenants can influence tar.gz artifacts processed by DBConnect artifact caching. The source bundle does not provide exact package ranges beyond before v3.7.0.
Exploitation context
KEV is false, and the provided sources do not state active exploitation. Exploitation requires control over a tar.gz file that MLflow extracts, plus user interaction per CVSS. The main concern is arbitrary file overwrite or sandbox escape in shared environments.
Researcher notes
Evidence is strongest for the vulnerable function, missing tar path validation, CWE-22/CWE-29 classification, and fixed threshold before v3.7.0. The bundle does not include proof of exploitation, exploit maturity, or complete downstream vendor impact.
Mitigation direction
- Upgrade mlflow/mlflow to v3.7.0 or later where feasible.
- Review the referenced MLflow commit and vendor advisories for exact remediation guidance.
- Restrict who can upload or supply archives processed by MLflow.
- Avoid processing untrusted tar.gz artifacts in shared or privileged runtime contexts.
- Apply least privilege to MLflow service accounts and filesystem permissions.
Validation and detection
- Inventory MLflow deployments and record installed versions.
- Identify workflows using DBConnect artifact caching or tar.gz artifact extraction.
- Confirm whether any deployment runs a version before v3.7.0.
- Check whether untrusted users can provide artifacts to affected workflows.
- Review filesystem permissions around MLflow runtime and cache directories.
Public sources used
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
CWE-22: File access and web shell behavior lookup
File traversal and upload weaknesses can lead teams to review file, web shell, execution, and collection telemetry. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
Open ATT&CK lookupCWE-29: Exact CWE lookup
Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
Open ATT&CK lookupFile access behavior lookup
The CVE wording references file access or upload behavior, so file telemetry and web shell review may help. This is a Glexia inferred lookup path, not an official MITRE, ATT&CK, or CVE Program mapping.
Open ATT&CK lookupCVE-2025-15036 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
Open ATT&CK lookup- Severity
- Critical
- CVSS
- 9.6 (3.0)
- Known Exploited
- No
- Published
Vector: CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
CNA and ADP enrichment extracted from CVE v5
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
SSVC decision data
CVSS vector scores
2 official scoresWe collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.
CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H2.86@huntr_aiCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H2.86redhat-SADPVulnerability scoring details
Base CVSS 3.1 score
9.6CriticalVector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
Base CVSS 3.0 score
9.6CriticalVector: CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
Vulnerability timeline
Timeline events are normalized from CVE metadata, CNA source timelines, ADP timelines, and KEV metadata when present.
- CVE reservedCVE Program
The CVE ID was reserved by the assigning CNA.
- ADP timelineredhat-SADP
Made public.
- CVE publishedCVE Program
The CVE record was published.
- ADP timelineredhat-SADP
Reported to Red Hat.
- CVE updatedCVE Program
The CVE record metadata indicates this as the latest update time.
ADP provider summaries
- 2026-03-30T02:01:17.426Z: Reported to Red Hat.
- 2026-03-30T01:16:06.400Z: Made public.
Source materials
- CVE List V5 sourceCVE List V5
- https://huntr.com/bounties/36c314cf-fd6e-4fb0-b9b0-1b47bcdf0eb0CVE reference, @huntr_ai
- https://github.com/mlflow/mlflow/commit/3bf6d81ac4d38654c8ff012dbd0c3e9f17e7e346CVE reference, @huntr_ai
- https://access.redhat.com/security/cve/CVE-2025-15036CVE reference, redhat-SADP · vdb-entry, x_refsource_REDHAT
- RHBZ#2452925CVE reference, redhat-SADP · issue-tracking, x_refsource_REDHAT
- https://security.access.redhat.com/data/csaf/v2/vex/2025/cve-2025-15036.jsonCVE reference, redhat-SADP · x_sadp-csaf-vex
Products and packages named in the record
CWE details
CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.
Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')
Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.
Path Traversal: '\..\filename'
Path Traversal: '\..\filename' represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.