@thsa2501 wrote:
.... why additional code need for fillet function? ....
Because, for example, the things you want to Fillet are the results from Offsetting the things that you actually draw. That means code would be required after each Offset to get information [such as endpoints] about the last object, and save it somewhere, and then after the next source Line is drawn and Offset, more code to get information about what is now the last object, so that the Fillet operation can use some location on each of those newer Lines. But it can't use just any point on them. In this situation [with the white being the ones you draw and the red being the results of Offsetting those]:

using just the start point or the end point of each red Line could result in any of these:

or it might even "see" one of the white Lines in Filleting. So code would be needed to find some unambiguous place on each red Line that is not on a white Line, such as its midpoint, and even that could lie on some other object that the Fillet command might see instead of the right one:

Have you tried MLINE yet? It will do what you want, with no custom command definition.
Kent Cooper, AIA