General guidelines for code generation
Use these general guidelines for code generation to get better code suggestions and create useful and accurate scripts.
Writing prompts
- Write clear and specific but concise prompts
- Specify the expected outcome and context clearly. Include important details like task requirements, specific APIs if you know them, and any limitations.
- Experiment with different prompts
- Consider adjusting your task instructions and incorporating examples. Observe how the code suggestions change with different prompt styles and levels of detail.
- Keep track of your prompts, along with any modifications and instructions for generating prompts that meet your specifications.
- Character limit of the prompts
-
Short and concise prompts generate better outcomes.
On reaching 200 characters, a message appears to inform you that short, focused, task-oriented directions yield the best results.
Input beyond 300 characters isn’t allowed.
| Weak prompt | Strong prompt | Notes |
|---|---|---|
| Get incidents with tasks | Get incidents with related tasks |
Includes sufficient detail. |
| Count P1 incidents between 3-3 and 4-13 | Use Glide aggregate to count number of P1 incidents closed between March 3 to April 13 assigned to admin |
Includes the API name and more specific language. |
| Don’t allow changing P1 change requests | If open change request is P1, don’t allow reducing the severity unless it's the creator |
Includes more specific instructions on what shouldn't change. |
| Latest change | Glide record of the most recent change |
Includes the API name and more specific language. |
Reviewing code
- Review code
- Implement strict and detailed reviews of the AI-generated code to determine its accuracy, efficiency, and how well it adheres to your coding standards.
- Test code
- Validate the code by running it against test cases in controlled environments to verify that it functions according to your requirements.