What you provide
- A meeting transcript, rough notes, or pasted discussion record
- The actual wording that supports decisions and commitments
- Optional instructions to return only selected sections
A focused Custom GPT that separates discussion, proposals, decisions, and explicit tasks while keeping unstated owners and due dates visible.
Runs inside ChatGPT. Sign-in and current ChatGPT plan availability may apply. No Store position or anonymous access is promised.

Illustrative input
Maya will send the revised outline by Friday. The team discussed the budget, but no decision was made.
Expected format
TL;DR
The revised outline has one explicit owner and due date. The budget remains unresolved.
Decisions
None stated
Action items
| Owner | Action | Due date | Evidence |
| --- | --- | --- | --- |
| Maya | Send the revised outline | Friday | "Maya will send the revised outline by Friday." |
Open questions
- What budget decision is still needed?Illustrative format only. The example demonstrates the evidence boundary rather than a live result.
Meeting material is sent to ChatGPT. Remove unnecessary personal details, customer information, credentials, legal discussion, and confidential company material before pasting when those details are not required.
It should not. Missing values are labeled [owner not stated] or [due date not stated] instead of being inferred.
Suggestions and proposals stay separate from decisions unless the supplied material explicitly records agreement or a decision.
Yes. One conversation starter extracts decisions, owners, and due dates, and you can request only the sections you need.
Only when appropriate for your organization and ChatGPT account. Remove unnecessary private or confidential details first.
Choose another input and output contract from the same verified page set.
Store-public readback: 2026-07-27
A narrow formula assistant that diagnoses one formula at a time, preserves locale and references, and uses placeholders when workbook details are missing.
A narrow Custom GPT for faithful transcription. It preserves visible structure, labels uncertainty, and returns the text before any explanation.