eRegs does not have any sort of background process which polls the Federal Register waiting for new data. Further, it does not currently handle eCFR data. Instead, it attempts to derive data from the final rules themselves. Unfortunately, this is a manually triggered, only semi-automated process requiring significant individual review.
Until we implement a better solution, however, it works, and allows us to display rules effective in the future. After a final rule has been published in the Federal Register, the parser should be re-ran and the appropriate XML file modified (if needed) to generate a correct set of changes. See the parser’s more thorough docs on this subject to learn more.