Claude to turn PTA votes into a reviewed report
Ei-chan, a working parent serving on the PTA, used Claude Cowork to prepare the results of the annual general meeting. The inputs were this year's Google Forms responses and last year's Word report, with the opening greeting already edited.
note.com
Original prompt
Shared by the source author. See source notes for context.
総会員数の欄だけは私が後で手入力するので空欄のままにしておいて。
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Ei-chan
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author reports an updated six-motion Word results report in roughly fifteen minutes, including corrected reply drafts and preserved formatting.
- Human involvement
- Ei-chan requested anonymization and wording corrections, retained membership entry and required human review of organizational replies.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Provide the current response CSV and the previous Word report.
- Aggregate votes and update the existing tables while leaving membership blank.
- Review anonymous reply drafts and enter the total membership manually.
Ei-chan, a working parent serving on the PTA, used Claude Cowork to prepare the results of the annual general meeting. The inputs were this year's Google Forms responses and last year's Word report, with the opening greeting already edited.
Cowork ran Python to split 215 submissions into 148 electronic responses and 67 proxy forms. It counted votes across six motions and updated the existing document rather than rebuilding its formatting. Last year's two-motion table was expanded to cover the six current motions, while the total membership field stayed blank for the author to enter.
The workflow also drafted responses to parents' written comments. Ei-chan gave specific corrections: anonymize names, preserve the original questions instead of paraphrasing them, shorten answers and keep the number of response items consistent. The article gives inconsistent counts for the written comments in different sections, so no precise comment count is repeated here.
Cowork reread paragraphs and tables before presenting the final document. Ei-chan reports approximately fifteen minutes for the work, compared with an estimate of several hours manually. The author still calls human review of proposed replies essential for accuracy and the PTA's position. The concrete artifact is the updated report with vote tables and response drafts, with a deliberately retained human-input field.
Prompt
Original prompt
Source: note.com · Ei-chan
Frequently asked questions
Why was membership left blank?
The author explicitly reserved the total-membership field for later manual entry.
Were parents' questions rewritten?
After feedback, Cowork preserved the CSV question wording and shortened the answer drafts.
Could the reply drafts be published unchecked?
The author says human review remains necessary for factual accuracy and the PTA's organizational position.
More details
About this account
Cowork executed vote aggregation, edited the Word structure and drafted responses to parent comments.
Firsthand source opened and read. Prompt is a verbatim excerpt of the published original instruction.
Credit: Ei-chan
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Cowork · Google Forms CSV · Word · Python
PTA voting · Word template · Parent feedback
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.