The libtcod renderers have specific limitations. Converting your renderer to fit with libtcod's contexts will remove many of the features that make your renderer worth using over libtcod's strict single-layer cell-based API. Because of this, I'd rather not treat your library like a libtcod context (rendering mode.)

I'd rather suggest your library as something which can be used as an alternative to libtcod's context. Especially now that libtcod can be compiled without SDL/GL dependencies.

Maybe you can add a new header which can render a libtcod console with the extra features of your renderer. A new header would make the libtcod integration optional instead of making your library dependent on libtcod or vice versa. This header can be held and maintained in libtcod's repo or your own.

I see what you mean. How about I create a renderer like this within libtcod and make a pull request?

You could, but I'm so busy trying to clean up libtcod. For me the best thing you can do is handle any integration yourself and add ports of your libraries to Vcpkg to make them more accessible.

I'm mostly hoping your library would be something like BearLibTerminal. A good alternative that I can suggest to anyone frustrated with libtcod's simplistic rendering, but independent of libtcod itself.