Microsoft Researcher Reveals How Three Chained Azure Flaws Enabled Cross-Tenant Identity Takeover

A case-sensitive regex check and a query string bypass were among three flaws chained to compromise any Azure Automation account’s managed identity, scoring CVSS 9.9.

Published on Aug 6, 2026
Joel Witts Written by Joel Witts
Microsoft Researcher Reveals How Three Chained Azure Flaws Enabled Cross-Tenant Identity Takeover

A Microsoft security researcher has revealed the full exploit chain behind CVE-2025-29827, a CVSS 9.9 vulnerability in Azure Automation that could have allowed attackers to steal managed identities across tenant boundaries.

Shay Shavit, a senior security researcher on Microsoft’s Azure Networking Security Research team, first reported the vulnerability last year. There have been no known exploits of the vulnerability in the wild.

At Black Hat 2026, Shavit shared the technical details publicly for the first time, walking through the three chained flaws that made the attack possible.

Azure Automation accounts are created with a public-facing endpoint by default. Shavit found that the service’s authorization check used the full URL, including query parameters, rather than just the path to determine whether a request required authorization. 

Appending &location to any request bypassed the check entirely, giving unauthorized access to certificates, credentials, and notebook content stored in any public automation account.

To reach the managed identity, which is typically highly privileged because automation accounts need broad access across Azure resources, Shavit needed to bypass a second layer. 

The service routed managed identity requests to a stricter authentication handler, but the routing logic used a case-sensitive regex match. Changing a single lowercase letter to uppercase caused the router to skip the strict handler while the downstream controller still processed the request normally.

“A capital N basically broke all of Azure’s authorization,” Shavit said. The result: any attacker with their own Azure account could query another tenant’s automation endpoint and obtain its managed identity token, gaining access to key vaults, storage accounts, and VM resources.

The vulnerability follows a similar pattern to a critical Entra ID flaw disclosed last year that allowed global admin impersonation across all tenants (CVE-2025-55241, CVSS 10.0), underscoring ongoing challenges with cross-tenant isolation in Microsoft cloud infrastructure.

Microsoft has patched all three flaws, reordered the authentication handlers, and added authorization checks at the controller level. 

Azure Automation accounts are still created with public endpoints by default, but Microsoft now recommends using private networking. Shavit urged organizations to audit their automation accounts for public exposure.

This field is for validation purposes and should be left unchanged.

FREE NEWSLETTER

Cyber Weekly

Get curated cybersecurity news, threats and insights delivered free every Thursday.

Written By Written By
Joel Witts
Joel Witts Content Director

Joel is the Director of Content and a co-founder at Expert Insights; a rapidly growing media company focused on covering cybersecurity solutions.

He’s an experienced journalist and editor with 8 years’ experience covering the cybersecurity space. He’s reviewed hundreds of cybersecurity solutions, interviewed hundreds of industry experts and produced dozens of industry reports read by thousands of CISOs and security professionals in topics like IAM, MFA, zero trust, email security, DevSecOps and more.

He also hosts the Expert Insights Podcast and co-writes the weekly newsletter, Decrypted. Joel is driven to share his team’s expertise with cybersecurity leaders to help them create more secure business foundations.