This vulnerability in pgAdmin 4's POST /browser/server/restore_point/{gid}/{sid} endpoint arises from improper neutralization of special elements in an SQL command (CWE-89). The 'value' field is interpolated directly into an SQL string using str.format() instead of being passed as a bound parameter, enabling SQL injection by authenticated users with an active PostgreSQL session. The injected SQL executes under the user's existing database role, which already has direct SQL access, so the impact is limited to the user's existing privileges. The fix involves passing the restore point name as a bound parameter and schema-qualifying the function call to pg_catalog.pg_create_restore_point to avoid function call hijacking via search_path manipulation. A regression test ensures the parameter is bound correctly. The vulnerability affects pgAdmin 4 versions from 1.0 before 9.16. No official patch or remediation level is stated in the vendor advisory.

An authenticated pgAdmin user with a connected PostgreSQL session can inject additional SQL statements via the named restore point endpoint. The injected SQL executes with the same privileges as the user's database role, which already has direct SQL execution capabilities through the Query Tool. Therefore, there is no privilege escalation or impact beyond the user's existing permissions. The vulnerability could allow SQL execution through an unexpected application path, potentially bypassing application-layer restrictions that gate the Query Tool.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. The described fix involves passing user input as bound parameters and schema-qualifying function calls to prevent SQL injection and function hijacking. Until an official fix is released, restrict access to authenticated users and monitor for unusual activity. Deployments that gate the Query Tool at the application layer should verify that this endpoint is similarly protected.