HighVulnerability

CVE-2026-71211: MLflow AI Gateway SSRF via unvalidated api_base in gateway secrets

Source: NVD(opens in a new tab)
Identifier: CVE-2026-71211
Published: August 5, 2026
Record updated: October 9, 2026

Affected: mlflow >= 3.13.0, <= 3.15.2
Fixed in: No fixed version was stated when the source was last read.
Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.3%

Summary

MLflow's AI Gateway stores an auth_config.api_base value verbatim when a gateway secret is created, with no scheme, host, or IP range validation. The raw_proxy endpoint in mlflow/server/gateway_api.py then sends requests to that stored address plus a caller-supplied path and returns the full response body. Because CreateGatewaySecret has no entry in the permission-validator map, any authenticated user, including read-only accounts, can exploit this, potentially exposing cloud-instance IAM credentials via metadata services. The flaw is independent of CVE-2026-4035.

Mitigation

The source does not state a fix yet. Check the original advisory for updates.

Details

Scores and exploitation

CVSS score: 7.1(high)
CVSS vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:L/A:N
Attack vector: network
Attack complexity: low
Privileges required: low
User interaction: none
EPSS: 0.3%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 4, 2026.
Disclosure date: August 5, 2026

Weaknesses and techniques

Weakness (CWE): CWE-918
Attack pattern (CAPEC): CAPEC-664

Affected

Packages: mlflow@>= 3.13.0, <= 3.15.2

Classification

Attack type: Other
Attack sophistication: Moderate
AI component targeted: API
Impact: Confidentiality
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative. The source gives this advisory a description and no title, so the title above is a headline generated from the source text.

Original source: https://nvd.nist.gov/vuln/detail/CVE-2026-71211

First tracked: August 5, 2026 at 08:08 AM ET, from NVD/CVE Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 92% (self-assessed)

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

mlflowPyPILLM dependency since 2024-05-30 · 17 tracked dependents

Summary

Mitigation

Affected packages in the Exposure Registry

Related items