Skip to main content

guardrails-detectors EUVDEUVD-2026-42927

| CVE-2026-15143 CRITICAL
Server-Side Request Forgery (SSRF) (CWE-918)
2026-07-10 redhat GHSA-34cr-c6hf-4xgc
9.3
CVSS 3.1 · Vendor: redhat
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Access
    Reach guardrails-detectors file_type endpoint
  2. Delivery
    Submit crafted XSD with external reference
  3. Exploit
    Parser resolves attacker URL (SSRF)
  4. Execution
    Server fetches metadata/internal service or file://
  5. Persist
    Leak cloud credentials or internal data
  6. Impact
    Exfiltrate to attacker

Vulnerability Assessment

Weakness
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V4 - API and Web Service URL allowlisting, SSRF prevention
MITRE ATT&CK

2 techniques across 2 tactics: Initial Access, Credential Access

The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up.

Exploitation Conditions

The concrete prerequisite is that the attacker can reach the guardrails-detectors service and submit input to the file_type content detector that accepts an arbitrary XSD/XML Schema string, which the detector then parses and resolves without disabling external references. Per the CVSS vector (AV:N/AC:L/PR:N/UI:N) this requires no authentication and no user interaction against the exposed detector endpoint. Limiting factors: the guardrails-detectors file_type detector must be deployed and network-reachable by the attacker (in RHOAI it is a guardrails component that may sit behind internal service boundaries rather than the public internet), and the impact of SSRF depends on what internal URLs, metadata services, or files are reachable from the server's network position and privileges.

Impact

The signals are largely aligned toward high priority but incomplete. The provided CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N, score 9.3) indicates remote, unauthenticated, low-complexity exploitation with scope change - consistent with SSRF that crosses a trust boundary into internal services. However, no EPSS score, no SSVC decision data, no KEV listing, and no POC were provided, so real-world exploitation likelihood cannot be quantified from the input; the absence of these should not be read as low risk, only as unmeasured. The confidentiality-heavy impact (C:H, I:L, A:N) matches an information-disclosure SSRF rather than direct RCE. Net: treat as a genuine high-priority patching candidate given the network-reachable, unauthenticated vector, while noting that active-exploitation status is unknown and EPSS is missing.

Exploit ScenarioDerived

An attacker with network access to a guardrails-detectors endpoint submits content to the file_type detector containing a crafted XSD whose schema reference points at the cloud instance-metadata service or an internal-only URL. The server resolves the reference and returns or leaks the response, letting the attacker read cloud credentials or reach services that are not exposed externally; a second variant uses a file:// reference to read local files. No authentication or user interaction is required per the CVSS vector, and no public POC is currently identified.

Act Now Prioritize within 24-48h. Why: CRITICAL severity.
No patch listed yet 0.3% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv1 – Jul 10, 2026

Server-side request forgery in the file_type content detector of guardrails-detectors (a component shipped with Red Hat OpenShift AI/RHOAI) allows a remote attacker to submit an arbitrary XML Schema Definition (XSD) that the parser resolves without restriction, coercing the server into fetching attacker-chosen URLs or reading local files. Because the CVSS vector is AV:N/PR:N/UI:N with a changed scope and high confidentiality impact (base score 9.3), an unauthenticated attacker can pivot into internal networks and harvest secrets such as cloud provider credentials. There is no public exploit identified at time of analysis, and the flaw is not listed in CISA KEV.

RemediationAI

No vendor-released patch version is identified at time of analysis - the provided references point to the Red Hat CVE page (https://access.redhat.com/security/cve/CVE-2026-15143) and Bugzilla 2498165, not to a tagged fixed release - so monitor those advisories and apply the RHOAI errata/updated guardrails-detectors image as soon as Red Hat publishes it. As compensating controls until then: disable or restrict the file_type content detector from accepting user-supplied XSD/XML schema input, and harden the XML/XSD parser to forbid external schema resolution and entity expansion (disable external schemaLocation, xs:import/include over the network, and file:// access), which prevents the SSRF primitive at the cost of rejecting schemas that legitimately reference external resources. Additionally, apply egress network controls on the detector pods - deny outbound access to cloud metadata endpoints (169.254.169.254), link-local, and internal RFC1918 ranges - accepting that this may break any legitimate outbound schema fetches, and restrict inbound access to the detector API to trusted callers only.

Affects your stack

Find out whether the stacks and servers you run are on an affected version.

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

The affected code is the file_type content detector within guardrails-detectors, the guardrails/input-validation tier of Red Hat OpenShift AI's LLM serving stack. The root cause is CWE-918 (Server-Side Request Forgery): when the detector validates or infers file type, it processes a user-supplied XSD schema, and XSD/XML schema processors will resolve external references (e.g. schemaLocation, xs:import/xs:include, or external entity-style URIs) unless explicitly hardened. Because those references are dereferenced by the server with the server's own network position and file-system privileges, an attacker controls the destination - arbitrary http(s) URLs (including cloud metadata endpoints like 169.254.169.254) or file:// paths - turning a schema-validation feature into a request-forgery and local-file-read primitive.

Know when Red Hat Openshift Ai Rhoai gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (redhat) PRIMARY 9.3 CRITICAL
AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N
vuln.today AI 9.3 CRITICAL

Remote unauthenticated XSD input (AV:N/AC:L/PR:N/UI:N); SSRF crosses trust boundary (S:C) reading credentials/files (C:H), minor internal state change (I:L), no availability impact.

3.1 AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N
4.0 AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:L/VA:N/SC:L/SI:N/SA:N
Red Hat 9.3 HIGH

Primary rating from Vendor (redhat).

CVSS VectorVendor: redhat

Attack Vector
Network
Attack Complexity
Low
Privileges Required
None
User Interaction
None
Scope
Changed
Confidentiality
High
Integrity
Low
Availability
None

DescriptionCVE.org

A flaw was found in the file_type content detector of guardrails-detectors. This vulnerability allows a remote attacker to supply an arbitrary XML Schema Definition (XSD) string, which is processed without proper restrictions. This can lead to server-side requests to arbitrary URLs or local file reads, potentially resulting in sensitive information disclosure, such as cloud provider credentials or access to internal network services.

Executive ViewAI

Why It Matters

Red Hat OpenShift AI's guardrails-detectors component contains a critical server-side request forgery vulnerability (CVE-2026-15143, CVSS 9.3) that allows unauthenticated attackers to exfiltrate secrets and pivot into internal infrastructure without authentication or user interaction. Organizations deploying OpenShift AI face immediate risk of cloud credential theft and lateral network compromise.

Business Impact: Severe
Compensating Controls

No vendor-released patch identified at time of analysis. Interim mitigations: restrict network egress from guardrails-detectors to block arbitrary URL fetching; implement WAF rules to detect and reject XSD payloads in requests; network-segment the detector from credential repositories and sensitive systems; disable the detector if operationally non-critical. Monitor Red Hat security advisories for patch release.

Threat Intelligence Overview

EPSS Score
0.3%
Top 83% exploitable
Prevalence
Low
niche guardrails component within Red Hat OpenShift AI, limited enterprise footprint
Social Media
1
Sightings
Data Quality
7 sources
Confidence: High
EUVD NVD CVE.org EPSS Vulnrichment Vendor CIRCL VulDB

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity CRITICAL
CISA KEV No
Public POC No
EPSS 0.3%
Patch available No
Active threat No
Deciding rule
CRITICAL severity
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision
Out-of-Cycle
Exploitation
None
System Exposure· derived
Open
Automatable
Yes
Human Impact
Very High
Technical Impact
Partial

Priority Score

46
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +46
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

Attack Technique

Server-Side Request Forgery exploits applications that fetch remote resources based on user-supplied URLs.

Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.

Affected Products (CPE)CVE.org

cpe:2.3:a:red_hat:red_hat_openshift_ai_(rhoai):*:*:*:*:*:*:*:*
cpe:2.3:a:red_hat:red_hat_openshift_ai_(rhoai):*:*:*:*:*:*:*:*
CVE-2026-23536 HIGH POC
7.5 Mar 20

The Feast Feature Server contains a path traversal vulnerability in its `/read-document` endpoint that allows unauthenti

CVE-2026-18948 CRITICAL
9.9 Aug 10

Unauthenticated remote code execution affects Feast (the open-source ML feature store), where user-defined functions sto

CVE-2026-14450 CRITICAL
9.9 Aug 10

Authentication bypass in the Models-as-a-Service (MaaS) API component of Red Hat OpenShift AI (RHOAI) lets any pod alrea

CVE-2026-5483 CRITICAL
9.9 Apr 10

Kubernetes Service Account token disclosure in the odh-dashboard component of Red Hat OpenShift AI (RHOAI) lets an authe

CVE-2026-23537 CRITICAL
9.1 Jul 01

Arbitrary file write in the Feast Feature Server's `/save-document` endpoint lets an unauthenticated remote attacker wri

CVE-2026-18951 HIGH
8.8 Aug 10

Privilege escalation in Red Hat OpenShift AI (RHOAI) training operator overlay allows any namespace editor to manage Tra

CVE-2026-18982 HIGH
8.8 Aug 10

Privilege escalation in Red Hat OpenShift AI's training-operator enables any user holding a standard Kubernetes edit or

CVE-2026-18617 HIGH
8.8 Aug 10

MySQL DSN parameter injection in Red Hat OpenShift AI's Data Science Pipelines Operator (DSPO) allows a namespace editor

CVE-2026-18950 HIGH
8.8 Aug 10

Privilege escalation in odh-dashboard, the web interface for Red Hat OpenShift AI (RHOAI), allows any authenticated dash

CVE-2026-18949 HIGH
8.8 Aug 10

Privilege escalation in odh-dashboard (Red Hat OpenShift AI) allows an attacker who has obtained the dashboard's Service

CVE-2026-13717 HIGH
8.8 Aug 10

The MaaS (Model-as-a-Service) Gateway component in Red Hat OpenShift AI (RHOAI) is improperly configured, enabling any l

CVE-2026-16745 HIGH
8.8 Jul 23

Authentication bypass in the odh-dashboard web console of Red Hat OpenShift AI (RHOAI) lets a low-privileged actor alrea

Vendor StatusVendor

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy