AI Threat Alert
CVE-2026-37067: Veno File Manager: unauth log disclosure via broken ACL
AWAITING NVDVeno File Manager 4.4.9 ships an admin endpoint, /vfm-admin/admin-panel/view/save-cvs.php, that fails to enforce authentication, letting anyone on the network send a crafted POST request and export the full application log history from a chosen date forward. There is no CVSS score, no EPSS data, no CISA KEV listing, and no public exploit or Nuclei template today, so urgency signals are thin, but the exploit itself needs zero privileges and zero AI/ML expertise — one unauthenticated request is enough, and package risk scoring shows no downstream dependents tracked. Veno File Manager is not one of our tracked AI/ML packages, and the 'ml_ui' tagging most likely reflects its use as a bundled file-browsing component in some data or AI ops stacks rather than a core model, agent, or inference risk — treat this as generic infrastructure exposure rather than an AI-specific attack primitive. Action: if this tool sits anywhere in your environment (including as a file browser inside an MLOps or data pipeline toolchain), pull the admin panel out of public reach immediately (network ACL or reverse-proxy auth in front of /vfm-admin/), watch for a vendor patch (none published in this advisory), and audit historical traffic to save-cvs.php for signs of prior log exfiltration.
What is the risk?
No CVSS/EPSS scoring is available and there is no evidence of active exploitation, a public PoC, or a scanner template, which keeps the near-term likelihood of mass exploitation low. However, the flaw is trivially exploitable — unauthenticated, single unsigned POST request, no chaining required — so any internet-facing deployment is realistically at risk from opportunistic scanning. Impact is bounded to information disclosure (application logs), not code execution or data modification, but logs can still contain file paths, usernames, IP addresses, or operational metadata useful for reconnaissance and follow-on attacks. Overall: low-to-moderate risk, driven up by trivial exploitability and down by the disclosure-only impact and lack of confirmed AI/ML system dependency.
How does the attack unfold?
What systems are affected?
| Package | Ecosystem | Vulnerable Range | Patched |
|---|---|---|---|
| n/a | — | — | No patch |
Do you use n/a? You're affected.
How severe is it?
What should I do?
1 step-
1) Restrict network access to /vfm-admin/ via firewall/WAF rules or reverse-proxy authentication so the endpoint is unreachable from the public internet. 2) Check the vendor (veno.com) and the GitHub advisory (jfs-jfs/CVE-2026-37067) for a patched release; none is documented in this record, so track for an update. 3) If Veno File Manager cannot be immediately isolated, disable the save-cvs.php export feature or add a compensating authentication layer at the web server level. 4) Review existing logs for unexpected POST requests to save-cvs.php to determine whether log data has already been exfiltrated, and rotate any credentials or tokens that may have appeared in exposed logs. 5) Inventory where this file manager is deployed across data/AI infrastructure, since it is not part of our standard AI package catalog and may be an unmanaged shadow-IT component.
How is it classified?
Which compliance frameworks are affected?
This CVE is relevant to:
Frequently Asked Questions
What is CVE-2026-37067?
Veno File Manager 4.4.9 ships an admin endpoint, /vfm-admin/admin-panel/view/save-cvs.php, that fails to enforce authentication, letting anyone on the network send a crafted POST request and export the full application log history from a chosen date forward. There is no CVSS score, no EPSS data, no CISA KEV listing, and no public exploit or Nuclei template today, so urgency signals are thin, but the exploit itself needs zero privileges and zero AI/ML expertise — one unauthenticated request is enough, and package risk scoring shows no downstream dependents tracked. Veno File Manager is not one of our tracked AI/ML packages, and the 'ml_ui' tagging most likely reflects its use as a bundled file-browsing component in some data or AI ops stacks rather than a core model, agent, or inference risk — treat this as generic infrastructure exposure rather than an AI-specific attack primitive. Action: if this tool sits anywhere in your environment (including as a file browser inside an MLOps or data pipeline toolchain), pull the admin panel out of public reach immediately (network ACL or reverse-proxy auth in front of /vfm-admin/), watch for a vendor patch (none published in this advisory), and audit historical traffic to save-cvs.php for signs of prior log exfiltration.
Is CVE-2026-37067 actively exploited?
No confirmed active exploitation of CVE-2026-37067 has been reported, but organizations should still patch proactively.
How to fix CVE-2026-37067?
1) Restrict network access to /vfm-admin/ via firewall/WAF rules or reverse-proxy authentication so the endpoint is unreachable from the public internet. 2) Check the vendor (veno.com) and the GitHub advisory (jfs-jfs/CVE-2026-37067) for a patched release; none is documented in this record, so track for an update. 3) If Veno File Manager cannot be immediately isolated, disable the save-cvs.php export feature or add a compensating authentication layer at the web server level. 4) Review existing logs for unexpected POST requests to save-cvs.php to determine whether log data has already been exfiltrated, and rotate any credentials or tokens that may have appeared in exposed logs. 5) Inventory where this file manager is deployed across data/AI infrastructure, since it is not part of our standard AI package catalog and may be an unmanaged shadow-IT component.
What systems are affected by CVE-2026-37067?
This vulnerability affects the following AI/ML architecture patterns: ML/AI-adjacent file management UI tooling, MLOps/data pipeline auxiliary infrastructure.
What is the CVSS score for CVE-2026-37067?
No CVSS score has been assigned yet.
What is the AI security impact?
Affected AI Architectures
MITRE ATLAS Techniques
AML.T0036 Data from Information Repositories AML.T0049 Exploit Public-Facing Application Compliance Controls Affected
What are the technical details?
Original Advisory
Incorrect access control in /vfm-admin/admin-panel/view/save-cvs.php in Veno File Manager Project 4.4.9 allows an unauthenticated attacker to extract all application logs from a desired date forwards via a specially crafted POST request.
Exploitation Scenario
An attacker scans for exposed Veno File Manager instances, identifies the unauthenticated admin panel, and sends a crafted POST request to save-cvs.php specifying a start date. The server responds with a CSV/log export containing the application's full activity history from that date onward — potentially revealing internal usernames, file paths (including paths to datasets or model artifacts if the tool is used in an AI/data pipeline context), and administrative actions. The attacker uses this information for reconnaissance, credential harvesting, or to identify additional attack surface within the broader environment, without needing any prior authentication or AI/ML-specific knowledge.
References
Timeline
Related Vulnerabilities
CVE-2025-5120 10.0 smolagents: sandbox escape enables unauthenticated RCE
Same attack type: Data Leakage CVE-2023-3765 10.0 MLflow: path traversal allows arbitrary file read
Same attack type: Data Leakage CVE-2025-2828 10.0 LangChain RequestsToolkit: SSRF exposes cloud metadata
Same attack type: Auth Bypass CVE-2025-53767 10.0 Azure OpenAI: SSRF EoP, no auth required (CVSS 10)
Same attack type: Auth Bypass CVE-2026-21858 10.0 n8n: Input Validation flaw enables exploitation
Same attack type: Auth Bypass