Critical LeakyLooker Vulnerabilities in Google Looker Studio Expose Cross-Tenant Data Risks
In a significant cybersecurity revelation, researchers have identified nine critical vulnerabilities within Google Looker Studio, collectively termed LeakyLooker. These flaws could have allowed malicious actors to execute unauthorized SQL queries across different organizational databases, leading to potential data breaches within Google Cloud environments.
Overview of the Vulnerabilities
The security firm Tenable brought these issues to light, emphasizing that while there is no evidence of these vulnerabilities being exploited in the wild, their existence posed a substantial risk. Google addressed these concerns following responsible disclosure in June 2025.
The identified vulnerabilities include:
1. Cross-Tenant Unauthorized Access via Zero-Click SQL Injection on Database Connectors: This flaw could have enabled attackers to inject malicious SQL commands without user interaction, compromising connected databases.
2. Cross-Tenant Unauthorized Access through Stored Credentials: Attackers could exploit stored credentials to gain unauthorized access across tenant boundaries.
3. Cross-Tenant SQL Injection on BigQuery via Native Functions: This vulnerability allowed SQL injection attacks specifically targeting BigQuery through its native functions.
4. Cross-Tenant Data Sources Leak via Hyperlinks: Sensitive data could be exposed through improperly secured hyperlinks, leading to unauthorized data access.
5. Cross-Tenant SQL Injection on Spanner and BigQuery via Custom Queries: Custom queries could be manipulated to perform SQL injection attacks on Spanner and BigQuery databases.
6. Cross-Tenant SQL Injection on BigQuery and Spanner via the Linking API: The Linking API could be exploited to execute unauthorized SQL commands across tenants.
7. Cross-Tenant Data Sources Leak via Image Rendering: Malicious actors could leverage image rendering processes to access sensitive data from other tenants.
8. Cross-Tenant XS Leak on Arbitrary Data Sources via Frame Counting and Timing Oracles: This sophisticated attack method could reveal data across tenants by analyzing frame counts and timing.
9. Cross-Tenant Denial of Wallet via BigQuery: Attackers could disrupt services by targeting BigQuery, leading to denial-of-service conditions.
Implications and Potential Exploits
Security researcher Liv Matan highlighted that these vulnerabilities compromised fundamental design principles, introducing a new class of attacks. They could have allowed attackers to exfiltrate, insert, or delete data within victims’ services and Google Cloud environments.
Organizations utilizing Google Sheets, BigQuery, Spanner, PostgreSQL, MySQL, Cloud Storage, and other Looker Studio data connectors were potentially at risk. Exploitation of these flaws could grant threat actors access to entire datasets and projects across different cloud tenants.
For instance, attackers could scan for public Looker Studio reports or gain access to private ones using connectors like BigQuery. By doing so, they could take control of databases, executing arbitrary SQL queries across the owner’s entire Google Cloud Platform (GCP) project.
In another scenario, if a victim created a public report or shared it with a specific recipient using a JDBC-connected data source such as PostgreSQL, attackers could exploit a logic flaw in the report’s copy feature. This flaw allowed cloning of reports while retaining the original owner’s credentials, enabling unauthorized deletion or modification of tables.
A particularly concerning exploit involved one-click data exfiltration. By sharing a specially crafted report, a victim’s browser could be forced to execute malicious code, contacting an attacker-controlled project to reconstruct entire databases from logs.
Conclusion
These vulnerabilities undermined the core security promise that a ‘Viewer’ should not have control over the data they are viewing. They could have enabled attackers to exfiltrate or modify data across Google services like BigQuery and Google Sheets. The prompt response and remediation by Google following the disclosure have mitigated these risks, but the incident underscores the importance of continuous vigilance and robust security practices in cloud environments.
