Trial 5: "Using AI to summarize, extract, and transcribe tasks"

This is an example of an AI-Studio trial. I reviewed previous trials.

Step 1: Use AI to summarize the description and enter it in a custom field.

  • When a task is added to Project 11, the AI ​​summarizes the description and enters it in the “Summary” custom field.
  • When a task is added to Project 12, the AI ​​summarizes the description and enters it in the “Summary” custom field.

Step 2: Use AI to extract task due dates by period and add them to the new task’s description.

  • In Project 21, extract tasks in Project 11 with due dates within a specified period (from today to five days from now) and enter the dates in the description of the new task in chronological order.
    At that time, enter the URL of the extracted task and the “Summary” custom field together. Click on the URL to open the original task. (AI does not re-summarize.)
  • In Project 22, extract tasks in Project 12 with due dates within a specified period (from today to three days ago) and enter the dates in the description of the new task in chronological order.
    At that time, enter the extracted task URL and the custom field “Summary” together. Click on the URL to open the original task (no AI re-summary will be performed).

Step 3: Use AI to extract task due dates by period and add them to the new task’s description.

  • In Project 3, extract tasks with today’s due date from Projects 21 and 22 and transcribe them into a new task.

If possible, I would like to process the above tasks all at once.
However, even after trying various AI guidance methods to process all the data at once, the AI was unable to produce the results we wanted.
Therefore, I have separated the process into three steps.
Each of the three steps requires the use of AI-Studio, which consumes three AI credits. However, since summaries are not performed multiple times, I think the consumption of AI credits can be reduced.


Reference: AI(Smart Workflow) credits used
AI Model = “GPT-5 mini”
Step 1: Approximately 400/operation x 5 operations/day = 4,000/day
Step 2: Approximately 600/operation x 2 operations/day = 1,200/day
Step 3: Approximately 600/operation x 1 operation/day = 600/day
Total AI Credits: Approximately 5,800/day

Currently, AI-Studio guidance allows to specify projects with @.
It would be good if could also specify sections with @.
What do you think?

For now, the AI ​​understands if you say, “Project section ‘AAAA’.”


Addendum
In Steps 2 and 3, the AI ​​further summarized the summary from Step 1. Summarization was performed a total of three times.
Therefore, we added the following to the AI guidance in Steps 2 and 3: “No summarization is necessary when transcribing.” It appears that no further summarization of the summary was performed after that.