I have Claude itself write CLAUDE.md. Once it is informed of its context (e.g., "README.md is for users, CLAUDE.md is for you") you can say things like, "update readme and claudemd" and it will do it. I find this especially useful for prompts like, "update claudemd to make absolutely certain that you check the API docs every single time before making assumptions about its behavior" — I don't need to know what magick spell will make that happen, just that it does happen.
Having been through cycles of manual writing with '#' and having it do it itself, it seems to have been a push on efficacy while spending less effort and getting less frustrated. Hard to quantify except to say that I've had great results with it. I appreciate the spirit of OP's, "CLAUDE.md is the highest leverage point of the harness, so avoid auto-generating it" but you can always ask Claude to tighten it up itself too.
Generally speaking it has a lot of information from things like OP's blog post on how best to structure the file and prompt itself and you can also (from within Claude Code) ask it to look at posts or Anthropic prompting best practices and adopt those to your own file.
This will start to break down after a while unless you have a small project, for reasons being described in the article.
Do you have any proof that AI written instructions are better than human ones? I don't see why an AI would have an innate understanding on how best to prompt itself.