Inspired by workspace-blueprint, I've built an "extension" of writing-room, which I use for linkedin posts captions. Wanted to have some consistent visual anchors under a "brand identity". post-designer takes a finalized caption, maps it to the best fitting preset supporting or visualizing the message, finds or designs icons, authors the title, labels, accents, and assembles in react canvas and outputs PNG. An important part for me is visual consistency through different post types, so it operates within a bunch of predefined rules and patterns. I am pretty happy with the results, but I have no idea how efficient this solution actually is :-D Maybe some kind of Figma/Canva mcp would get me there faster. On the other hand, that would be a potential 3rd party cost, and I would miss an opportunity to f**k around to find out ^^.