Tend to adopt RFC standards for specs, works pretty well. My general rule of thumb is describe scope/concerns/interfaces at a high level, details at any wanted depth. That makes the LLM generated code reasonably stable and bounds it to a clearly defined boundary. Whatever ambiguity is left in details is then treated "expectedly unstable".
Comments
Tend to adopt RFC standards for specs, works pretty well. My general rule of thumb is describe scope/concerns/interfaces at a high level, details at any wanted depth. That makes the LLM generated code reasonably stable and bounds it to a clearly defined boundary. Whatever ambiguity is left in details is then treated "expectedly unstable".
Do you have an example of such a spec that you can publicly share, perhaps on GitHub?