What is Prompt Engineering?
Prompt Engineering is the art and science of writing effective prompts to get the most out of generative AI. A good prompt guides the AI to understand the context and come up with more accurate, relevant, and useful responses. Structuring a prompt correctly can make the difference between a generic result and a highly optimized response for your use case.Why is it important?
Michelangelo.land is designed to work in perfect harmony with your development flow, but the quality of its answers also depends on how the prompt is formulated. Through intelligent use of Prompt Engineering, you can:- Get more precise code suggestions.
- Generate more relevant content.
- Reduce the need for manual iterations.
- Customize the output of Michelangelo.land to your needs.
The Prompt Engineering Section
This section of the documentation collects a list of strategies and best practices to improve the quality of your prompts and make the most of Michelangelo.land. Here you will find:- Fundamental principles for writing effective prompts.
- Advanced techniques to obtain more precise results.
- Practical examples of prompts optimized for different use cases.
- Common mistakes to avoid to maximize the quality of responses.
Prompting Strategies
Depending on your specific needs, these strategies can be very effective when used together. We encourage you to try them out to see which ones yield the best results. While Michelangelo.land can already do a lot with a basic, generic prompt, using some of these techniques can help you get even better results.Contextual prompting
Providing context can help Michelangelo.land better understand the scope of your requests.There are two ways to provide context directly:
- by clicking the “+Add content” button to tag files
- including context directly in the prompt itself by setting the stage with basic information, for example:
Setting the context:
Incremental prompting
Making gradual, smaller changes produces better results.Avoid ambiguity
Make sure your prompts are clear and unambiguous. Avoid vague terms and be as specific as possible about what you need.Defining Project Structure and File Management
Clearly outline your project structure, such as filenames and paths, to ensure organized and manageable code generation.Define the project structure:
Debugging and Issue Reporting
When reporting bugs or requesting changes be specific. Report the issue, describe the current behavior, describe the expected behavior, and add constraints.Describe the issue:
Using Console Log for reporting bugs
Pasting an error logged in the Log Console into the chat can be very helpful.Report the log: