CVE-2026-118173: MISP Guard Incomplete Tag Validation Allows Non-Allowed Taxonomy Tags to Bypass Feed Policy Enforcement
MISP Guard, a proxy that enforces security policies on MISP feed data, contained an incomplete validation check for required taxonomy tags. When a feed event or its attributes carried tags belonging to a required taxonomy (for example, TLP), the guard only verified that at least one tag of that taxonomy matched the configured allowed list. It did not verify that every tag of the required taxonomy present on the event or attribute was in the allowed list. As a result, an event or attribute could simultaneously carry an allowed tag (e.g., tlp:clear) and a non-allowed tag (e.g., tlp:red or tlp:amber) and still pass the guard's policy check. This allowed data marked with a higher classification or sensitivity level to be distributed through feeds configured to only distribute lower-classification data, effectively bypassing the intended data-handling restrictions. The vulnerability affected both event-level and attribute-level tag validation, including attributes nested within objects and shadow attributes. An attacker with the ability to author or modify MISP events in a feed source could include non-allowed taxonomy tags alongside allowed ones to exfiltrate more sensitive data to feed consumers than the guard policy intended to permit.
Affected Software
Event History
Frequently Asked Questions
Who could exploit this issue?
An attacker needs the ability to author or modify MISP events in a feed source. They can attach an allowed tag and a non-allowed tag from the same required taxonomy so the event or attribute passes the guard's policy check.
What content can bypass the feed policy?
Both event-level and attribute-level tags are affected. This includes attributes nested within objects and shadow attributes, when they carry both an allowed and non-allowed tag from a required taxonomy.
What is the practical impact of a successful bypass?
Data carrying a higher classification or sensitivity tag can be distributed through feeds intended to allow only lower-classification data. For example, an item tagged both tlp:clear and tlp:red or tlp:amber could pass a policy allowing tlp:clear.