Reverse Prompt Engineering Multiple Tweets

I want to use reverse prompt engineering, where you help me create prompts based on tweets I give you that would be optimized and ideal for producing similar tweets.
can you give me an example of generalized reverse prompt engineering a tweet using multiple examples? After drawing inspiration from 3 tweets show a prompt that starts with "write a tweet"
Combine them and extract the most semantically relevant entities from these {tweet}. Put the entities into a chart, numbering them in the first column, listing them in the second column, and including their salience scores in the third column. tweet = [tweet] tweet = [tweet2] tweet = [tweet3]
Make a reverse prompt engineer the {tweet}. Give me a single generalized prompt that would create a similar tweet. The tone and writing style should be included in the prompt. The entities #1-2 should also be mentioned as needing to be included in the output in the prompt. Ignore entities #3-4. Leave entity numbers out of the prompt.
===== copy generated reverse prompt to new ChatGPT chat =====
much shorter