The best way to learn how to use AI effectively is to experiment and explore what works best. Prompts are how you tell your AI what to do. Clear, detailed prompts will improve the quality of your results. Great prompting is all about:
  1. Communicating your desired result as clearly as possible. Providing examples or defining what it means for the task to be “done” are both great techniques.
  2. Providing context. Remember, you can @-mention files from your workspace, @-mention specific tools you want Zo to use, or share relevant URLs from the internet.
  3. Splitting large tasks into a “plan” composed of smaller components.
If you want Zo to do a large task, try asking Zo to come up with a plan first.We recommend asking Zo to write the plan to a note – this will make it easier for you to review and edit the plan (yourself, or with Zo’s help), before asking Zo to execute the plan.
You can learn more about how the Zo team uses AI to build Zo in our blog post: https://www.zo.computer/blog/model-selection We’ll keep extending this blog post as we learn more and improve our techniques. We’ve accumulated many hard-earned learnings and best practices from working with AI over the last few years, and we’ve baked many of our principles into the Zo AI system itself. And of course, we often use Zo to build parts of Zo.