Misskey CVE-2026-46712

Exploitation requires a valid, authenticated user account (PR:L) on the target Misskey instance running version 2025.3.2 through 2026.5.3. The CVSS 4.0 AT:P metric indicates a specific attack requirement is present - most likely the attacker must possess or enumerate valid DM thread or message object identifiers belonging to other users. Exploitation is bounded to the Direct Messages (Chat) feature; the 'specified' visibility note type is explicitly not affected. Federation does not need to be enabled. Impact is limited to read access of DM data points (VC:L); there is no write, modification, or denial-of-service potential. Attackers outside the platform (unauthenticated) cannot exploit this vulnerability.

Real-world risk is low across all available signals. The CVSS 4.0 score of 2.3 reflects network reachability (AV:N) but is constrained by specific attack requirements (AT:P), low-privilege authentication (PR:L), and limited confidentiality impact (VC:L) with no integrity or availability impact. AT:P in CVSS 4.0 indicates exploitation depends on conditions that are not universally present - likely knowledge of target DM thread or message identifiers. No EPSS data was provided; given the narrow scope (DM data points only, no lateral impact), exploitation probability is expected to be low. There is no CISA KEV listing and no known public POC, confirming this is not a priority for threat-led response. The primary risk is unauthorized read access to private conversation metadata or content within a single Misskey instance, limited to authenticated users. This is a genuine low-severity issue: limited blast radius, authentication barrier, no write or availability impact, and no evidence of active exploitation.

An authenticated Misskey user submits a crafted API request to the DM (Chat) endpoint, supplying a manipulated object identifier (consistent with CWE-639) that references a thread or message belonging to another user. The server fails to verify the requesting account's permission to access that object and returns the data. Because AT:P applies, the attacker likely needs prior knowledge of valid DM identifiers - obtained through enumeration or information leakage - before successfully extracting unauthorized message data. No public POC exists at the time of analysis.