The tool, which allows users to add context notes to potentially misleading posts, went public last month after a pilot program in the US.