Critical Exploits Target MLflow and FUXA Vulnerabilities in August 2026
Executive Summary
In August 2026, two critical vulnerabilities were actively exploited: CVE-2026-64849 in MLflow and CVE-2026-25895 in FUXA. The MLflow vulnerability allowed unauthenticated attackers to perform Server-Side Request Forgery (SSRF) attacks, enabling access to internal cloud metadata endpoints and extraction of sensitive data. The FUXA vulnerability permitted unauthenticated remote attackers to write arbitrary files to the server filesystem, potentially leading to remote code execution. Both vulnerabilities were promptly patched in subsequent software releases.
The exploitation of these vulnerabilities underscores the persistent targeting of open-source platforms by threat actors. Organizations are urged to prioritize timely patching, conduct thorough audits for signs of compromise, and implement robust security measures to protect against similar threats.
Why This Matters Now
The active exploitation of these vulnerabilities highlights the urgent need for organizations to update their MLflow and FUXA installations to the latest versions to prevent unauthorized access and potential system compromises.
Attack Path Analysis
Attackers exploited an unauthenticated SSRF vulnerability in MLflow to access internal cloud metadata endpoints, extracting sensitive credentials. With these credentials, they escalated privileges within the cloud environment. Subsequently, they moved laterally to other cloud services and resources. The attackers established command and control channels to maintain persistent access. They exfiltrated sensitive data, including cloud credentials and secrets. Finally, they potentially disrupted services or deployed malicious payloads to achieve their objectives.
Kill Chain Progression
This analysis maps confirmed threat intelligence to the full cloud kill chain to show where defensive gaps would emerge as an attack progresses.
Initial Compromise
Description
Attackers exploited an unauthenticated SSRF vulnerability in MLflow to access internal cloud metadata endpoints, extracting sensitive credentials.
Related CVEs
CVE-2026-25895
CVSS 9.8A path traversal vulnerability in FUXA allows an unauthenticated, remote attacker to write arbitrary files to arbitrary locations on the server filesystem, potentially leading to remote code execution.
Affected Products:
frangoteam FUXA – <= 1.2.9
Exploit Status:
active scanning observed
MITRE ATT&CK® Techniques
Exploit Public-Facing Application
Valid Accounts
OS Credential Dumping
Exfiltration Over Alternative Protocol
Ingress Tool Transfer
Command and Scripting Interpreter
Application Layer Protocol
Remote Services
Potential Compliance Exposure
Mapping incident impact across multiple compliance frameworks.
PCI DSS 4.0 – Ensure all system components and software are protected from known vulnerabilities
Control ID: 6.2
NYDFS 23 NYCRR 500 – Cybersecurity Policy
Control ID: 500.03
DORA – ICT Risk Management Framework
Control ID: Article 5
CISA ZTMM 2.0 – Asset Management
Control ID: 3.1
NIS2 Directive – Cybersecurity Risk Management Measures
Control ID: Article 21
Sector Implications
Industry-specific impact of the vulnerabilities, including operational, regulatory, and cloud security risks.
Information Technology/IT
MLflow SSRF vulnerability enables cloud credential theft from AI/ML platforms, requiring immediate patching and security review of exposed instances.
Industrial Automation
FUXA SCADA/HMI exploitation allows remote code execution on operational technology systems, threatening critical manufacturing and control infrastructure security.
Oil/Energy/Solar/Greentech
Critical infrastructure sectors face severe operational disruption from FUXA vulnerabilities targeting SCADA systems controlling energy production and distribution networks.
Health Care / Life Sciences
AI platform vulnerabilities compromise patient data protection and research systems, violating HIPAA compliance through unauthorized cloud credential access.
Sources
- Attackers Exploit MLflow SSRF Flaw to Steal Cloud Credentials and Secretshttps://thehackernews.com/2026/08/attackers-exploit-mlflow-ssrf-flaw-to.htmlVerified
- NVD - CVE-2026-25895https://nvd.nist.gov/vuln/detail/CVE-2026-25895Verified
- FUXA Security Advisory GHSA-88qh-cphv-996chttps://github.com/frangoteam/FUXA/security/advisories/GHSA-88qh-cphv-996cVerified
- FUXA Release Notes v1.2.10https://github.com/frangoteam/FUXA/releases/tag/v1.2.10Verified
Frequently Asked Questions
Cloud Native Security Fabric Mitigations and ControlsCNSF
Based on the attack progression modeled above, these are the defensive controls that would constrain each stage.
Aviatrix Zero Trust CNSF is pertinent to this incident as it would likely limit the attacker's ability to move laterally and exfiltrate data by enforcing strict segmentation and controlled egress policies.
Control: Cloud Native Security Fabric (CNSF)
Mitigation: The attacker's ability to exploit the SSRF vulnerability may have been constrained, reducing the likelihood of unauthorized access to internal metadata endpoints.
Control: Zero Trust Segmentation
Mitigation: The attacker's ability to escalate privileges may have been limited, reducing the scope of unauthorized access within the cloud environment.
Control: East-West Traffic Security
Mitigation: The attacker's lateral movement could have been constrained, limiting their ability to access additional cloud services and resources.
Control: Multicloud Visibility & Control
Mitigation: The attacker's ability to establish and maintain command and control channels may have been reduced, limiting persistent access.
Control: Egress Security & Policy Enforcement
Mitigation: The attacker's ability to exfiltrate sensitive data could have been limited, reducing the risk of data loss.
The attacker's ability to disrupt services or deploy malicious payloads may have been constrained, reducing the potential impact on cloud operations.
Impact at a Glance
Affected Business Functions
- SCADA Operations
- Industrial Automation Control
Estimated downtime: 3 days
Estimated loss: $50,000
Potential exposure of operational technology configurations and control data.
Recommended Actions
Key Takeaways & Next Steps
- • Implement Zero Trust Segmentation to restrict unauthorized access and limit lateral movement within the cloud environment.
- • Deploy Egress Security & Policy Enforcement to monitor and control outbound traffic, preventing unauthorized data exfiltration.
- • Utilize Multicloud Visibility & Control to detect and respond to anomalous activities across cloud services.
- • Apply Inline IPS (Suricata) to identify and block known exploit patterns and malicious payloads.
- • Regularly update and patch MLflow and other critical systems to mitigate known vulnerabilities.
