Large Language Model Cue Word Optimization Tool

Optimize your cue words, just like PromptPerfect, to tap into the potential of large language models for more accurate and relevant responses.

silver lining: Provides a powerful template to help you create accurate, effective cue words to guide the model in generating quality responses.

 

The key idea in optimizing prompts is to stick to the fundamental rules that were in place when the prompt was written:

A cue command usually includes the following elements:

- Instructions: Explicit tasks or specific directions that you expect the AI model to perform.
- Context example: cover any external information or additional details that could help the AI generate a response more accurately.
- Input Data: A specific query or input to which you would like an answer.
- Output indication: indicates the format or type of response expected.

Usage:

- If you are calling LLM's API directly, you can enter it as system and user.
- If the question is asked directly with chatgpt/gpt4, it can be entered in two parts

Example 1

Before：

```
Classify the text into neutral, negative or positive. 
```

After:

```
Prompt: <Analyze the sentiment of the following text snippet and categorize it as either 'neutral', 'negative', or 'positive'. Please provide a brief justification for your classification to offer insight into your reasoning process.>
```

Example 2

Before:

```
Table departments, columns = [DepartmentId, DepartmentName]
Table students, columns = [DepartmentId, StudentId, StudentName]
Create a MySQL query for all students in the Computer Science Department
```

After:

```
Write a MySQL query to retrieve all records of students who are part of the Computer Science Department. Assume there are two tables involved: one named 'departments' with columns 'DepartmentId' and 'DepartmentName', and another named 'students' with columns 'DepartmentId', 'StudentId', and 'StudentName'. The 'DepartmentId' column in the 'students' table references the 'DepartmentId' in the 'departments' table as a foreign key. The query should list all students where 'DepartmentName' is "Computer Science". Ensure that the query is syntactically correct and properly formatted for use in a MySQL database.
```

Example 3

Before:

```
translate the following English content into Chinese, 并润色
```

After:

```
<Please translate the following English text into Chinese, ensuring not only accuracy in translation but also enhancing the prose to flow naturally and elegantly in Chinese. The translation should read as if originally written by a native speaker, with attention to cultural nuances and idiomatic expressions.>
```

Example 4

Before:

```
完善句子
```

After:

```
请根据以下的中文句子开头，完善每个句子。你的句子补全应当语法正确，上下文恰当，并且展现出一定的创造性或有趣的思考，使句子吸引人。请确保遵循每个句子开头提供的风格或语气指示。
```