Apache Superset: Sensitive Data Exposure via REST API (disabled by default)
Description
A Sensitive Data Exposure vulnerability exists in Apache Superset allowing authenticated users to retrieve sensitive user information. The Tag endpoint (disabled by default) allows users to retrieve a list of objects associated with a specific tag. When these associated objects include Users, the API response improperly serializes and returns sensitive fields, including password hashes (pbkdf2), email addresses, and login statistics. This vulnerability allows authenticated users with low privileges (e.g., Gamma role) to view sensitive authentication data
This issue affects Apache Superset: before 6.0.0.
Users are recommended to upgrade to version 6.0.0, which fixes the issue or make sure TAGGING_SYSTEM is False (Apache Superset current default)
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
CVE-2026-23983 exposes sensitive user data (password hashes, emails) via the Tag endpoint in Apache Superset before 6.0.0, even for low-privilege users.
Vulnerability
Description
CVE-2026-23983 is a sensitive data exposure vulnerability in Apache Superset affecting versions before 6.0.0 [1][3]. The root cause lies in the Tag endpoint, which is disabled by default but, when enabled, allows authenticated users to retrieve objects associated with a tag [1]. If those objects include User entities, the API response improperly serializes and returns sensitive authentication data, specifically PBKDF2 password hashes, email addresses, and login statistics [1][3].
Exploitation
Prerequisites
Exploitation requires the attacker to be an authenticated user, including those with the low-privilege Gamma role [1][3]. The feature must be explicitly enabled by an administrator (TAGGING_SYSTEM set to True), as it is disabled by default [1][3]. An attacker would craft a request to the Tag endpoint for a tag associated with user objects, thereby receiving the serialized sensitive fields [1].
Impact
If exploited, an attacker gains access to password hashes (PBKDF2), which could be cracked offline, as well as email addresses and login statistics [1][3]. This could lead to account takeover or further credential-based attacks, depending on the privilege level exposed. The confidentiality of user authentication data is compromised, and the issue is rated with a CVSS score (not fully detailed but noted as a security risk) [1].
Mitigation
Apache Superset has fixed the issue in version 6.0.0 [1][3]. Users are strongly recommended to upgrade immediately. As an alternative, administrators should ensure the TAGGING_SYSTEM configuration remains set to False (the default) to disable the vulnerable endpoint entirely [1][3].
AI Insight generated on May 19, 2026. Synthesized from this CVE's description and the cited reference URLs; citations are validated against the source bundle.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
apache-supersetPyPI | < 6.0.0 | 6.0.0 |
Affected products
2- Apache Software Foundation/Apache Supersetv5Range: 0.0.0
Patches
0No patches discovered yet.
Vulnerability mechanics
AI mechanics synthesis has not run for this CVE yet.
References
4- github.com/advisories/GHSA-h294-8fxm-m2pjghsaADVISORY
- lists.apache.org/thread/62mgbc5hc8026skp69kb6vqozj3pr5wwghsavendor-advisoryWEB
- nvd.nist.gov/vuln/detail/CVE-2026-23983ghsaADVISORY
- www.openwall.com/lists/oss-security/2026/02/24/7ghsaWEB
News mentions
0No linked articles in our index yet.