I've been doing a version of this in a side project. Instead of saving the prompt directly, I have a road map. When implementing features, I tell it to brainstorm implementation for the road map. When fixing a bug, I tell it to brainstorm fixes from the roadmap. There's some back and forth, and then it writes a slice that is committed. Then, I look it over, verify scope, and it makes a plan (also committed). Then it generates work logs as it codes.

My prompts are literally "brainstorm next slice" or "brainstorm how to fix this bug" or "talk me through trades offs of approach A Vs B" so those prompts aren't meaningful in their own.

It's quite effective, but I'm a team of one.