Ghost CMS CVE-2026-105681

Requires a valid authenticated Ghost member account (PR:L) on an instance running an affected version (5.9.0 up to but not including 6.44.1) with the native comments feature in use. No non-default configuration is needed beyond having comments enabled. Exploitation is limited by the need for member-level authentication (not fully anonymous) and by scope: only comment data is exposed - there is no write/modify capability and no impact to posts, member PII stores, or site availability. Sites that do not use Ghost's comments feature are not meaningfully exposed.

This is a moderate-severity, authorization/input-validation flaw (CWE-943, NoSQL injection logic) in the Ghost comments subsystem, not a top-tier emergency. The provided CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N = 6.5) indicates remote, low-complexity exploitation that requires authentication (PR:L) - consistent with the description stating 'members' are the threat actors, so a valid member account is a genuine prerequisite. Impact is confined to confidentiality (C:H, I:N, A:N): unauthorized read of comments, with no write or availability effect. No EPSS score was supplied, the CVE is NOT on the CISA KEV list, and no public exploit (POC) is indicated - so exploitation status is 'no public exploit identified at time of analysis.' The source-code evidence corroborates the root cause: the comment-replies endpoint previously set frame.options.isAdmin = true before calling the generic read path, and post/filter scoping was applied inconsistently, letting a member reach admin-level/other-scope comment data; the fix introduces a dedicated adminRead path, centralizes post-scoping via #withPostFilter, and rejects unsupported count.reports filters. Net: a real but bounded data-exposure issue worth patching promptly on multi-tenant/membership sites, rather than a widely-weaponized critical. Note the C:H rating is arguably generous for comment-only data - see assessed vector.

An attacker registers or already holds a low-privileged Ghost member account on a site with comments enabled, then sends crafted requests to the comment-replies or comments filter endpoints to retrieve comments (including replies or reported/admin-scoped comment data) belonging to other members or posts they should not see. Because the CVSS vector is AV:N/AC:L/PR:L, the attack is remote and technically simple once a member session exists; no public exploit code is identified, but the fix diff clearly reveals the vulnerable logic.