A critical, unauthenticated SQL injection zero-day vulnerability, tracked as CVE-2026-72898, is being actively exploited in the popular open-source business intelligence platform Metabase. The vulnerability, which has a CVSS score of 10.0, affects Metabase versions 1.58 and later. It allows a remote, unauthenticated attacker to gain full administrator privileges on a vulnerable instance. Several organizations, including laptop manufacturer Framework, have already confirmed data breaches resulting from the exploitation of this flaw. Metabase has released patches and is urging all self-hosted customers to upgrade immediately or apply mitigations to prevent compromise.
The vulnerability is an unauthenticated SQL injection flaw located in the /api/session/reset_password API endpoint. An attacker can send a specially crafted POST request to this endpoint, injecting malicious SQL commands into the backend Metabase application database. Successful exploitation allows the attacker to effectively reset the password of an existing user, typically targeting an administrator account, and take it over.
Once an attacker has administrator access to the Metabase instance, they have complete control. They can:
This vulnerability is particularly dangerous because Metabase is designed to be a central hub for connecting to and visualizing data from numerous sensitive sources like production databases, data warehouses, and application backends.
Metabase Cloud instances were automatically patched by the vendor and are not vulnerable.
The vulnerability is confirmed to be actively exploited in the wild as a zero-day. Metabase themselves discovered the flaw after their own cloud infrastructure was targeted. Since the public disclosure on August 6, 2026, exploitation has become more widespread. Several companies have publicly disclosed that they were victims of attacks leveraging this vulnerability, including:
The impact of exploiting CVE-2026-72898 is catastrophic for affected organizations, justifying its CVSS 10.0 score. Gaining administrator access to Metabase is equivalent to handing over the keys to an organization's most critical data assets. The primary impact is widespread data breach, as seen in the Framework incident where customer PII (names, emails, phone numbers, addresses) was stolen. Attackers can pivot from the compromised Metabase instance to attack the underlying connected databases, potentially leading to further system compromise, data destruction, or ransomware deployment. The reputational damage and potential regulatory fines resulting from such a breach are significant.
The following patterns may help identify vulnerable or compromised systems:
Metabase provided a specific log pattern to identify compromise attempts:
POST request to /api/session/reset_password that returns a 400 status code.GET request to /api/user/current that returns a 200 status code.This sequence indicates an attacker successfully used the exploit to gain a session token and then verified their access.
/api/session/reset_passwordPOST /api/session/reset_password HTTP/1.1" 400GET /api/user/current HTTP/1.1" 200Log Analysis (D3-NTA): The most effective detection method is to analyze Metabase and web server access logs for the specific pattern described above. Create a SIEM alert that triggers when a POST to /api/session/reset_password with a 400 status is followed within a short time window by a GET to /api/user/current with a 200 status from the same source IP.
# Example Sigma Rule Logic
title: Metabase SQLi Zero-Day CVE-2026-72898 Exploitation
status: experimental
logsource:
category: webserver
detection:
selection_attempt:
cs-method: POST
c-uri: '/api/session/reset_password'
sc-status: 400
selection_success_check:
cs-method: GET
c-uri: '/api/user/current'
sc-status: 200
timeframe: 10s
condition: selection_attempt and selection_success_check | near(selection_attempt.source_ip == selection_success_check.source_ip)
Network Monitoring: Monitor network traffic for connections to the /api/session/reset_password endpoint. While legitimate password resets will use this, a high volume of requests or requests from known malicious IPs should be investigated.
/api/session/reset_password endpoint at your firewall or web application firewall (WAF). This will prevent exploitation but will also break the legitimate password reset functionality.Critical Metabase SQLi (CVE-2026-72898) added to CISA KEV catalog. New technical details on HoneySQL injection confirm full admin access and credential theft.
Immediately upgrade Metabase to a patched version to eliminate the vulnerability.
Mapped D3FEND Techniques:
As a temporary measure, use a WAF or firewall to block access to the vulnerable /api/session/reset_password endpoint.
Mapped D3FEND Techniques:
The most critical and effective countermeasure is to immediately upgrade all self-hosted Metabase instances to a patched version. Due to the active exploitation and CVSS 10.0 severity of CVE-2026-72898, this should be treated as an emergency change. Prioritize internet-facing instances first, followed by internal ones. Before updating, take a snapshot or backup of the instance for rollback purposes. After the update, verify that the application is running correctly and monitor logs closely for any continued signs of compromise, as attackers may have established persistence prior to the patch. Automate patch notifications for critical open-source software like Metabase to ensure your security team is alerted to future vulnerabilities promptly.
If patching is delayed for any reason, implement an emergency filtering rule on your Web Application Firewall (WAF) or reverse proxy to block all requests to the /api/session/reset_password endpoint. This is a temporary but effective mitigation that breaks the exploit chain. Be aware that this will also disable the legitimate password reset functionality for users. The rule should be highly specific to block POST requests to this exact path. This action buys time for a proper patching and incident response cycle. Once the Metabase instance is patched, this rule can be removed. This technique serves as a crucial compensating control when immediate patching is not feasible.
Proactively and retroactively analyze web server access logs for the specific attack signature: a POST to /api/session/reset_password returning status 400, followed by a GET to /api/user/current returning status 200 from the same IP. Implement this logic as a real-time alert in your SIEM. Hunt for this pattern in historical logs going back to the beginning of August 2026 to identify if you were compromised before the public disclosure. If a compromise is found, use the source IP and timestamps to pivot your investigation, examine all subsequent activity from that IP, and trigger your incident response plan, assuming all data accessible by Metabase has been exfiltrated.
Metabase discovers and discloses the zero-day vulnerability after being targeted.

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.
CyberNetSec.io uses automation to assist source monitoring, deduplication, observable extraction, and structured intelligence generation. Published analysis follows human-defined editorial standards and adds defensive context including MITRE ATT&CK, D3FEND, STIX, and Sigma where applicable. Read our editorial policy.
Help others stay informed about cybersecurity threats
Every tactic, technique, and sub-technique used in this threat has been identified and mapped to the MITRE ATT&CK framework for consistent, actionable threat language.
Observables and indicators of compromise (IOCs) have been extracted and cataloged. Risk has been assessed and correlated with known threat actors and historical campaigns.
Detection rules, incident response steps, and D3FEND-aligned mitigation strategies are included so your team can act on this intelligence immediately.
Structured threat data is packaged as a STIX 2.1 bundle and can be visualized as an interactive graph — relationships between actors, malware, techniques, and indicators.
Sigma detection rules are derived from the threat techniques in this article and can be converted for deployment across any major SIEM or EDR platform.