CVE-2026-104908 in MISP
Summary
by MITRE • 10/02/2026
MISP contains an improper input validation vulnerability in the decaying model import functionality. The import endpoint was intended to create a new decaying model belonging exclusively to the importing user's organisation, with the default flag forced to off.
However, the application stripped only the top-level id and uuid fields and pinned org_id and default on the outer array before saving the data flat. A user with decaying-model permissions could supply a nested model key carrying its own primary key, organisation identifier, and default flag, which bypassed those guards during the save operation.
Impact:
- A user with perm_decaying could overwrite an existing decaying model belonging to another organisation in place, altering its name, formula, parameters, or ownership.
- A user could create or modify a model flagged as the organisation default, affecting scoring behaviour for other users.
- A user could reassign a model's organisation to an arbitrary value.
Preconditions:
- Authenticated user with decaying-model permission (perm_decaying).
- Network access to the MISP instance.
Affected: <2.5.48.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 10/02/2026
The vulnerability identified in the Management Information System for Sharing, commonly known as MISP, represents a critical failure in input validation and object ownership enforcement within its decaying model import functionality. Decaying models are specialized data structures used to adjust the relevance of indicators over time based on configurable formulas and parameters. The security flaw stems from an improper handling of nested JSON objects during the import process. Specifically, while the application logic attempts to enforce organizational boundaries by stripping top-level identifiers such as id and uuid, and explicitly setting the outer array's org_id and default flags, it fails to recursively validate or sanitize these attributes within nested model definitions. This architectural oversight allows an authenticated attacker with decaying-model permissions to inject a maliciously crafted payload containing a nested key that carries its own primary key, organization identifier, and default flag settings. Because the save operation processes this data flat without re-applying organizational constraints to the inner objects, the application accepts these overrides as legitimate configuration changes rather than rejecting them as unauthorized modifications.
From an operational perspective, this vulnerability enables severe integrity violations within the threat intelligence sharing ecosystem. An attacker can overwrite existing decaying models that belong to other organizations in place, effectively hijacking their analytical logic by altering names, formulas, or parameters without detection. This capability extends beyond simple data corruption; it allows for the reassignment of model ownership to arbitrary organization identifiers, potentially masking malicious activity as legitimate organizational updates. Furthermore, an attacker can force a newly created or modified model to be flagged as the default decaying model for another organization. Since many MISP instances rely on these default models for automated scoring and prioritization of incoming indicators, this manipulation directly impacts the accuracy and reliability of threat assessments across the entire community relying on that instance. The impact is not limited to data integrity but extends to operational security, as incorrect decay calculations can lead to false positives or missed detections in downstream security operations centers utilizing MISP feeds.
This vulnerability aligns with CWE-270, which describes a privilege context switch error where an application fails to properly maintain the separation between different user contexts during state changes. Additionally, it maps to ATT&CK technique T1565.002, specifically Data Manipulation via Stored Payloads, as the attacker modifies stored data structures that are subsequently executed or utilized by other users and automated systems. The precondition for exploitation requires only an authenticated account with decaying-model permissions and network access to the MISP instance, making it a relatively low-barrier attack vector for insider threats or compromised accounts within participating organizations.
Mitigation strategies must focus on enforcing strict object ownership validation at every level of data structure processing rather than relying solely on top-level field manipulation. Developers should implement recursive sanitization that ensures all nested objects inherit the importing user's organizational context and are stripped of any externally provided identifiers, keys, or flags that could influence storage behavior. Input validation routines must verify that no part of the imported payload attempts to override ownership attributes before persisting data to the database. For organizations currently running versions prior to 2.5.48, immediate upgrading is required as this issue has been addressed in subsequent releases. Until an upgrade can be performed, administrators should restrict decaying-model permissions to a minimal set of trusted users and monitor import logs for anomalies involving nested object structures or unexpected changes to default model configurations.