/

CVE-2026-19193

Severity:

HIGH RISK

(7.8)

Improper Privileges (CWE-266) in Jiangmin Minifilter

CVE-2026-19193

Deferred

Vulnerability type

Category:

Improper Privilege Management

CWE:

CWE-266, CWE-284

Impact:

Local Privilege Escalation and Unauthorized Access

Severity:

HIGH RISK

(7.8)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Jiangmin Antivirus 21 contains an improper privilege management flaw in the MessageNotifyCallback function of the kvcore.sys Minifilter Port component. A local attacker may be able to bypass intended access controls and perform actions with privileges that should not be available to them.

Affected context

The issue occurs in Jiangmin Antivirus 21 where the kvcore.sys Minifilter Port processes messages through the MessageNotifyCallback function. Risk is greatest when a local user can reach the affected driver interface and the callback does not adequately verify authorization, object ownership, or the privilege level required for the requested operation.

Exploit preconditions

  • Attack surface: Local interaction with the antivirus kernel driver or its exposed minifilter communication interface
  • Required access: Local user access to the affected system; the required baseline privileges depend on the driver's access controls
  • Trigger point: Processing of messages by the MessageNotifyCallback function in kvcore.sys
  • Why exploitation is possible: Access-control checks may improperly authorize operations or fail to enforce the privilege boundary between untrusted local callers and kernel-level functionality

Why this happens

The underlying issue is improper privilege management, mapped primarily to CWE-266 and related to CWE-284: Improper Access Control. The driver appears to make security-sensitive decisions without consistently tying requested operations to the caller's identity, privilege level, or authority over the affected resource. This insecure pattern can allow a less-privileged local process to access functionality intended only for trusted components.

Potential impact

  • A local attacker may obtain unauthorized access to protected driver functionality
  • The flaw could enable local privilege escalation
  • An attacker may bypass security controls enforced by the antivirus component
  • System confidentiality, integrity, or availability can be affected
  • The vulnerability could support follow-on compromise of the host

General mitigation guidance

Mitigation generally focuses on strengthening privilege boundaries around kernel-driver interfaces and minimizing trust in local callers.

Common defensive patterns include:

  • Enforcing explicit caller authentication and authorization for every security-sensitive request
  • Applying least-privilege access to device objects, communication ports, and related resources
  • Validating requested operations against the caller's effective privilege and resource ownership
  • Separating privileged operations from ordinary message handling
  • Failing closed when identity, authorization, or request context cannot be established
  • Reducing the exposed attack surface of kernel components and restricting unnecessary local access
  • Using defense-in-depth controls such as protected services, driver isolation, and endpoint policy enforcement where appropriate

How teams detect issues like this

Organizations can look for this weakness through:

  • Static analysis of driver callbacks and IOCTL or minifilter communication handlers for missing authorization checks
  • Code review signals such as trust in caller-supplied identities, privilege checks performed only during initialization, or security decisions based solely on message fields
  • Testing with users of different integrity and privilege levels to verify that access decisions remain consistent
  • Monitoring endpoint telemetry for unexpected local processes interacting with antivirus driver interfaces
  • Policy enforcement that limits access to kernel-device objects and flags unauthorized driver communication

Automated security review tools may help identify exposed driver interfaces and incomplete access-control paths, but findings should be validated against the driver's intended security model.

Sources

SourceLink
github.comgithub.com/patrickt2017/CVE-2026-19193-PoC
vuldb.comvuldb.com/cve/CVE-2026-19193
vuldb.comvuldb.com/submit/864551
vuldb.comvuldb.com/vuln/386711
vuldb.comvuldb.com/vuln/386711/cti

Frequently asked

questions

The described attack requires local access to the affected system. Remote exploitation would generally require a separate mechanism that first provides local execution or access to the vulnerable driver interface.

An attacker needs a local account or process capable of reaching the relevant Minifilter Port interface. The exact baseline privilege requirement depends on how the driver exposes and protects that interface.

The safest pattern is to enforce authorization at every security-sensitive callback or request boundary, use least-privilege access to driver interfaces, and reject requests that cannot be associated with an adequately privileged caller.

Apply threat modeling to kernel communication paths, review all privilege transitions, test interfaces under multiple user contexts, and use static analysis and security policies to detect missing or inconsistent authorization checks.

Find out what's already
exploitable in your codebase.