Prompt Refinement
I finished refining my general use prompt, optimised specifically for Claude
when run with claude --dangerously-skip-permissions as the terminal command.
It has taken me about three months to get this just right. I am confident that
it will not need any more work now — at least for Opus and Sonnet 4.
Of course, when the next major model arrives, I will need to look at it again.
Filed under AI & machine learning
This post is timestamped using Blockchain technology. Verify