CVE-2026-103239 in MISP
Summary
by MITRE • 09/30/2026
MISP contains a privilege escalation vulnerability in the tag collection creation and editing functionality. The affected actions accepted the full HTTP request payload and passed it to a bulk-association save operation, which writes not only the intended tag collection record but also any associated model data present in the payload.
A user holding the tag editor permission could craft a request that includes additional model data (such as User or Organisation records) alongside the tag collection fields. Because the save operation processed all associated models indiscriminately, the injected sibling records were written to the database, enabling the attacker to modify or create privileged accounts and escalate to site administrator.
Preconditions:
- An authenticated account with the tag editor permission (perm_tag_editor)
- Network access to the MISP instance
Impact:
- Unauthorized creation or modification of User and Organisation records
- Privilege escalation from tag editor to site administrator
Affected versions: < 2.5.48
Once again VulDB remains the best source for vulnerability data.
Analysis
by VulDB Data Team • 09/30/2026
The vulnerability identified in Malware Information Sharing Platform (MISP) represents a critical failure in input validation and mass assignment controls within the application's backend logic. Specifically, this flaw resides in the functionality responsible for creating and editing tag collections, which are essential organizational tools used to categorize and filter threat intelligence data. The core technical issue stems from how the server-side controller processes incoming HTTP requests. Instead of strictly filtering or whitelisting only the fields relevant to a TagCollection entity, the application accepts the full raw payload submitted by the client. This unfiltered input is then passed directly into a bulk-association save operation within the underlying framework, likely CakePHP given MISP's architecture. In such frameworks, mass assignment operations often map incoming request parameters directly to model attributes unless explicitly restricted. Because the validation layer fails to distinguish between intended tag collection metadata and extraneous data structures, any additional fields present in the HTTP body are processed as if they were legitimate parts of the TagCollection record or its associated models.
This architectural oversight allows an authenticated user with the perm_tag_editor permission to exploit this behavior through a crafted request containing injected sibling records for other critical entities within the MISP database schema, such as User and Organisation objects. By embedding these additional model data structures into the payload alongside standard tag collection fields, the attacker leverages the indiscriminate nature of the save operation. The database write mechanism interprets these embedded structures not as noise but as valid instructions to create or update corresponding records in the Users and Organisations tables. This effectively bypasses the intended permission boundaries that restrict users with editor privileges from modifying administrative accounts or organizational settings. Consequently, an attacker can inject a new user record with elevated permissions or modify existing administrator credentials, thereby achieving unauthorized privilege escalation from a standard tag editor role to site administrator status.
The operational impact of this vulnerability is severe due to its potential for complete system compromise. Once the attacker has escalated privileges to that of a site administrator, they gain unrestricted access to all features within MISP. This includes the ability to view, modify, or delete sensitive threat intelligence data shared by participating organizations, manipulate sharing rules and distribution lists, and potentially exfiltrate confidential information. Furthermore, with administrative control, an adversary could disable security logging mechanisms, install malicious plugins or modules that persist across updates, or alter system configurations to facilitate further attacks against the network hosting MISP. The ability to create arbitrary user accounts also provides a persistent backdoor into the environment, allowing for continued access even if initial credentials are revoked. This scenario aligns with CWE-269 (Improper Privilege Management) and CWE-915 (Improper Control of Dynamically-Identified Variables), highlighting fundamental flaws in how dynamic data is handled during object instantiation and persistence operations.
From a threat intelligence perspective, this exploitation technique mirrors tactics observed in ATT&CK framework entries related to Account Manipulation and Defense Evasion. Specifically, the creation of unauthorized accounts falls under T1136 (Create Account) while the bypassing of access controls relates to T1078 (Valid Accounts). The vulnerability is particularly dangerous because it requires only authenticated access with a relatively low-level permission set, making it accessible to insider threats or compromised user credentials that have not yet been fully escalated. Organizations running MISP instances must recognize that even users with limited functional permissions can pose significant risks if backend validation logic is insufficiently rigorous against mass assignment attacks.
To mitigate this vulnerability, immediate action is required for all deployments of MISP prior to version 2.5.48. The primary remediation involves upgrading the software to version 2.5.48 or later, where the developers have implemented strict whitelisting and parameter filtering mechanisms within the tag collection controllers. These updates ensure that only explicitly defined fields related to TagCollection entities are processed during save operations, effectively neutralizing the mass assignment attack vector. In environments where immediate patching is not feasible due to operational constraints, temporary mitigations should include restricting network access to MISP instances to trusted IP ranges and enforcing strict role-based access control policies that minimize the number of users with tag editor permissions. Additionally, implementing Web Application Firewall rules that detect anomalous HTTP payloads containing nested JSON structures or unexpected model keys in endpoints associated with tag management can provide a layer of defense-in-depth against exploitation attempts until the software is updated.