⚠️ This column is a sharing of the prompt creation process for free educational purposes, and is not any investment, legal, or financial advice. AI execution results may contain errors and independent verification is essential, and all responsibility from utilization is entirely attributed to the user themselves. Cannot be directly used for business decision-making. [View all required confirmations below the 30-week timeline before use[]](https://www.notion.so/30-29f84ddcec3c80a1a87eda9883d81453?pvs=21)
In Week 7, I completed Strategy_Checklist.md. It was a systematic checklist of 1,000 lines or more, containing rules in PART A, search strategy in PART B, results record in PART C, and evaluation criteria in PART D. I managed it with text-editor's create, view, str_replace, insert commands, and automated the workflow by combining it with Sequential Thinking.
Now it was the turn to collect the second information type, target items. Among the 6 information types identified in Week 4, target items came after strategy classification. Strategy classification was the framework of the policy, and target items were the specific products within it.
However, I had a concern. Should I create a checklist for target item collection from scratch? Or could I refer to Strategy_Checklist.md?
I decided to refer to Strategy_Checklist.md. I opened the file with text-editor's view command and requested structure identification using Sequential Thinking. Claude instantly grasped the structure and characteristics, especially the PART A, B, C, D structure was clear. PART A was project information and rules, PART B was source-specific search strategy, PART C was results record, PART D was evaluation criteria.
The reason this structure was effective was because it matched the work order. First read the rules, check the search strategy, record the results, and finally evaluate. The order was clear.
I decided to use the same structure for Product_Checklist.md as well. Create PART A, B, C, D, but change the content to fit target items. I created an empty file with text-editor create command and saved it with the name "Product_Checklist.md", and started filling the content.
While writing PART A, I discovered the first difference. Strategy and Product had different search targets. Strategy_Checklist.md found the documents themselves. "Is this document Investigation, Proposed, or Final?" One document was one search result. I checked document title, publication date, status, and it was relatively simple. Completed when finding 60-75 documents.