I mentioned recently in a post about needing a Copilot for Copilots. Obviously, I’m not the first person to suggest this, and most times it’s said in jest, but in reality, this is more and more becoming a necessity for the reasons I laid out in that original post.
I didn’t note in that post, but I’m also big braining this and have been for some time. To help generate the best results we need the best prompts. We have to retrain ourselves in some respects to go from terse search terms to in-person style conversations to get the most out of modern tools infused with Generative AI. But with
I lay these out publicly so that you can a) watch the progress, and b) you can get involved with feedback, thoughts, ideas, and direct development. I’m always happy to get help getting to the goal faster.
The GitHub repo is: https://github.com/rod-trent/Prompt_Refiner
I have stacks of plans and diagrams for the ultimate vision and will share those as progress is made. My goal is to make this capability just part of the process. I envision a day where the tool will always produce the best prompt based on several sources of information and based on the context of the environment the tool is enacted from. I know that sounds a bit vague, but it’s really not. You’ll see.