The vulnerability lies in the OIDC token endpoint's authorization code validation logic. The advisory explicitly points out the flawed conditional statement in backend/internal/service/oidc_service.go. The code used a logical AND (&&) where a logical OR (||) was required. This meant that an authorization code was rejected only if both the client ID was incorrect and the code was expired. The correct behavior is to reject the code if either of these conditions is true. I confirmed the fix by finding the commit that changed the && to ||. The commit 34890235ba8c2d856e3a121fdf59fe9d627e8596 contains this exact change in the createTokenFromAuthorizationCode function within OidcService. Therefore, during exploitation, this function would be present in the runtime profile as it processes the malicious token exchange request.