Get the latest tech news
Show HN: Prompting LLMs in Bash scripts
Of course it's that easy.
I’ve been experimenting with using LLMs locally for generating datasets to test Harper against. I might write a blog post about the technique (which I am grandiosely calling“LLM-assisted fuzzing”), but I’m going to make you wait. For example, you can swap out the system prompt, so if you want to compare behavior across prompts, you can just toss it in a loop:
Or read this on Hacker News