Automated Threat Modeling

Brief Summary: Azure Cloud Shell CVE-2026-35428 Command Injection Leading to Network Spoofing (CVSS 9.6)

CVE-2026-35428 is a critical command injection vulnerability in Azure Cloud Shell, with a CVSS score of 9.6, that enables network spoofing. Microsoft has already deployed a server-side fix, and customers do not need to take any action. This short review covers the patch details and the affected systems.

CVE Analysis

6 min read

ZeroPath CVE Analysis
ZeroPath CVE Analysis

2026-05-07

Brief Summary: Azure Cloud Shell CVE-2026-35428 Command Injection Leading to Network Spoofing (CVSS 9.6)
Experimental AI-Generated Content

This CVE analysis is an experimental publication that is completely AI-generated. The content may contain errors or inaccuracies and is subject to change as more information becomes available. We are continuously refining our process.

If you have feedback, questions, or notice any errors, please reach out to us.

[email protected]

Introduction

A critical command injection flaw in Azure Cloud Shell, scored at CVSS 9.6, could have allowed an unauthorized attacker to perform network spoofing by exploiting how the service processes user-supplied input. The remediation model is as notable as the severity. Microsoft patched the vulnerability entirely on the server side before public disclosure, so no Azure customer needs to take any action.

This CVE is part of Microsoft's transparency initiative for cloud service vulnerabilities, announced in June 2024. Under this policy, Microsoft now assigns CVE IDs to critical cloud service flaws even when the fix has already been silently deployed. The exclusively-hosted-service tag on the CVE.org record marks this classification, so defenders and security tools can tell vulnerabilities that need customer remediation apart from those the cloud provider has already handled.

Technical Information

Vulnerability Classification

CVE-2026-35428 is classified under CWE-77: Improper Neutralization of Special Elements used in a Command (Command Injection). The vulnerability is in Azure Cloud Shell, an interactive, authenticated, browser-accessible terminal for managing Azure resources. Cloud Shell sessions run on temporary hosts provided on a per-user basis, with a 20-minute idle timeout and a 5 GB file share for persistent storage.

Root Cause

The flaw is in how Azure Cloud Shell handles user-supplied input. Special elements in commands were not sufficiently neutralized, which allowed an attacker to inject malicious content and, in turn, to spoof over a network. Neither Microsoft nor any public disclosure has published detailed exploit mechanics. The CWE-77 classification still tells us that the service failed to properly sanitize or escape special characters before it incorporated user input into the commands it built on the backend.

CVSS 3.1 Breakdown

The full CVSS 3.1 vector string is: AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

MetricValueOperational Meaning
Attack VectorNetworkThe vulnerable component is bound to the network stack.
Attack ComplexityLowExploitation does not require advanced conditions.
Privileges RequiredNoneThe attacker does not need prior authentication.
User InteractionRequiredA user must interact for the exploit to trigger.
ScopeChangedThe vulnerability impacts resources beyond the vulnerable component.
ConfidentialityHighTotal loss of information confidentiality.
IntegrityHighTotal compromise of system integrity.
AvailabilityHighTotal loss of availability.

Attack Flow

From the available information, the attack flow is as follows:

  1. Initial Setup: The attacker crafts a malicious payload that exploits the command injection flaw in Azure Cloud Shell. The attacker does not need prior authentication to the target environment (Privileges Required: None).

  2. User Interaction: The attack requires some form of user interaction (UI:R). This likely means the attacker must lure a legitimate Azure Cloud Shell user into triggering the injected command, perhaps with a crafted link or a social-engineering lure that gets the victim to interact with malicious input in their Cloud Shell session.

  3. Command Injection: Once the user interacts, the Cloud Shell backend processes the improperly neutralized input and executes the injected command elements.

  4. Spoofing and Scope Change: The injected command enables spoofing over the network. The "Changed" scope means the impact extends beyond the Cloud Shell component itself and may reach other Azure resources or services accessible from the compromised session context. The High ratings for confidentiality, integrity, and availability suggest the attacker could read sensitive data, modify resources, and disrupt service availability.

Public disclosures do not describe the exploit mechanics beyond this, so the precise injection point and payload structure cannot be independently verified.

Patch Information

Microsoft has fully resolved CVE-2026-35428 with a server-side fix applied directly to the Azure Cloud Shell infrastructure. No customer action is required. The MSRC advisory, published on May 7, 2026, says so explicitly: "This vulnerability has already been fully mitigated by Microsoft. There is no action for users of this service to take."

Azure Cloud Shell is an exclusively hosted service that Microsoft manages entirely, so the patch went to the backend environment with no customer-downloadable updates, KB articles, or build number changes. The MSRC Security Update Guide reflects this. Its security updates table lists "Customer Action Required: Not Required" for the single affected product entry (Azure Cloud Shell). The CVSS temporal metric for Remediation Level is set to "Official Fix (O)", which confirms that a complete vendor solution is in place.

If you use Azure Cloud Shell, the flaw has already been neutralized at the platform level. There are no patches to download, no configuration changes to make, and no upgrade steps to follow. Microsoft handled the full remediation lifecycle internally before public disclosure.

Organizations should still consider these validation steps:

  • Ensure that Cloud Shell usage is restricted to trusted entry points such as the official Azure portal and Visual Studio Code extensions.
  • Monitor for unusual administrative actions or unexpected Cloud Shell session starts around the May 7, 2026 publication date.
  • Review Azure activity logs for anomalous behavior in Cloud Shell sessions during the window before the fix was deployed.

Affected Systems and Versions

The affected product is Azure Cloud Shell. Because it is an exclusively hosted cloud service, traditional version numbers, build identifiers, and downloadable software versions do not apply. The CVE.org record and the NVD listing do not specify affected version ranges, which is consistent with a server-side service.

Anyone who used Azure Cloud Shell before Microsoft's server-side remediation (deployed before the May 7, 2026 public disclosure) was potentially affected. The fix has been applied across the entire Azure Cloud Shell infrastructure, so all current sessions run on the patched environment.

Vendor Security History

The Microsoft Security Response Center investigates all reports of security vulnerabilities affecting Microsoft products and services. Its deployment of a server-side fix for this cloud service vulnerability, with no customer intervention required, shows a mature incident response capability. This CVE falls under the broader transparency initiative Microsoft announced in June 2024, under which the company now assigns CVE IDs to critical cloud service flaws even when customers do not need to apply patches. The policy is meant to make visible the vulnerabilities that were previously fixed silently on the server side, so defenders and security tools have a fuller picture of the threats they face.

References

Detect & fix
what others miss

Works with
  • GitHub
  • GitLab
  • Bitbucket
  • Azure DevOps Services
  • Gerrit
  • CVS
  • Jira
  • Linear
  • Slack
  • Security Compass
Explore integrations
Security magnifying glass visualization