everything AI writes sounds the same. someone made a markdown file format for giving an agent an actual personality

the tell with every autonomous agent is the same. output is competent, voice is completely generic. and the two fixes both suck: finetune a model on your own writing (expensive, slow, locked to one vendor) or paste "write in a casual tone" into a system prompt, which makes you sound exactly like everyone else who typed that sentence.

aeon's take on it is a thing called soul.md, basically a personality spec in plain markdown. identity and worldview, real opinions including contradictory ones (which is the part i think actually matters, real people hold inconsistent views), a separate style.md for sentence rhythm and vocabulary, a memory.md so it carries across sessions, and example outputs to anchor it.

clip's them building one around elon as a test case, which is a funny way to prove it works since you'd notice immediately if it didn't.

what gets me is it's portable. it's markdown, so it isn't tied to a model or a vendor, you just hand the same file to whatever you're running.