Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB
CVE-2026-111445 Overview
CVE-2026-111445 is a broken access control vulnerability in the Wheels framework, affecting versions 4.0.0 through 4.1.2. The flaw resides in findByKey() and exists(key=), where incoming scope conditions are replaced with a primary-key-only condition. Applications relying on scopes for ownership or tenant authorization can load, modify, or delete records belonging to other users or tenants when the key originates from a request. The issue is tracked under [CWE-669: Incorrect Resource Transfer Between Spheres].
Critical Impact
Authenticated attackers can read, update, or delete records outside their authorized tenant or ownership scope by supplying another entity's primary key to vulnerable finder methods.
Affected Products
- Wheels 4.0.0 through 4.1.2
- Fixed in Wheels 4.2.0
- Applications using findByKey(), exists(key=), updateByKey(), or deleteByKey() with scope-based authorization
Discovery Timeline
- 2026-10-16 - CVE-2026-111445 published to NVD
- 2026-10-19 - Last updated in NVD database
Technical Details for CVE-2026-111445
Vulnerability Analysis
The Wheels Object-Relational Mapping (ORM) layer provides scope mechanisms that developers commonly use to enforce tenant isolation or record ownership. In the affected releases, findByKey() and exists(key=) discard any caller-supplied scope conditions and query solely by primary key. A record outside the caller's authorized scope is returned as if it belonged to the caller.
The behavior of exists(key=) is particularly dangerous in production. When error information is disabled, the method returns true for an out-of-scope key instead of throwing an exception. Application code that treats this boolean as an authorization check will then proceed to operate on the referenced record.
Downstream calls to updateByKey() and deleteByKey() inherit the same flaw, enabling cross-tenant modification or deletion. Other scope-aware finders, aggregates, updateAll(), and deleteAll() are not affected.
Root Cause
The root cause is improper composition of query predicates. The vulnerable methods overwrite the WHERE clause produced by scopes with a primary-key equality condition rather than combining the two with a logical AND. This collapses the authorization boundary into a direct key lookup.
Attack Vector
An authenticated attacker sends a request containing a primary key value belonging to another tenant or user. The application calls findByKey(), exists(key=), updateByKey(), or deleteByKey() with that key, trusting the scope to enforce ownership. The scope is silently dropped, the record is returned, and the attacker proceeds to view or mutate data they do not own.
// Example exploitation pattern (prose description)
// Vulnerable application code:
// record = model("invoice").findByKey(key=params.id, where="tenantId=#session.tenantId#")
// The `where` and any model scope are discarded; the row for params.id is returned
// regardless of tenantId. See the GitHub Security Advisory for details.
Detect Identity Attacks the Moment Authentication Fails
Singularity™ Identity surfaces credential misuse, lateral movement, and privilege escalation in real time, so you can contain the breach before it spreads.
Stop Identity AttacksDetection Methods for CVE-2026-111445
Indicators of Compromise
- Application logs showing successful reads, updates, or deletes on records whose owner or tenant identifier does not match the authenticated session.
- Unexpected HTTP 200 responses for GET, PUT, or DELETE requests referencing primary keys that fall outside a user's dataset.
- Audit trails revealing updateByKey() or deleteByKey() operations preceded by exists(key=) checks against foreign keys.
Detection Strategies
- Perform code review across the application for calls to findByKey(), exists(key=), updateByKey(), and deleteByKey() that pass user-controlled key values.
- Instrument the data access layer to log the final SQL predicate and flag queries missing expected tenant or owner conditions.
- Add integration tests that submit another user's primary keys to protected endpoints and assert authorization denial.
Monitoring Recommendations
- Alert on sequences where a session accesses a monotonically increasing set of record identifiers it has not previously interacted with.
- Monitor web application firewall (WAF) telemetry for key enumeration patterns against endpoints backed by scope-based authorization.
- Correlate database audit logs with application identity context to surface cross-tenant row access.
How to Mitigate CVE-2026-111445
Immediate Actions Required
- Upgrade Wheels to version 4.2.0 or later, which combines key conditions with incoming scope, query-builder, and where conditions in the affected finders.
- Audit controllers and services for reliance on exists(key=) as an authorization gate and add explicit ownership checks.
- Rotate or invalidate any records whose integrity cannot be confirmed following triage.
Patch Information
The fix is included in Wheels 4.2.0. The patched methods — findByKey(), exists(key=), updateByKey(), and deleteByKey() — now compose primary-key predicates with caller-supplied scope, query-builder, and where conditions using logical AND. Refer to the GitHub Security Advisory GHSA-2qh9-qcxq-f8qc for the full advisory.
Workarounds
- Replace vulnerable findByKey() calls with scope-aware finders such as findOne() that accept explicit where conditions and are not affected.
- Perform ownership verification in application code before invoking updateByKey() or deleteByKey(), comparing the loaded record's tenant or owner identifier to the session.
- Avoid exposing raw primary keys in client-facing identifiers; use indirect references scoped to the user's session.
# Upgrade Wheels to the fixed release
box install wheels@4.2.0
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB

