Meeting notes to decisions, action items and follow-up message
Turns raw meeting notes or a transcript into a clean list of decisions, action items with owners and dates, open questions, and a ready-to-send recap. Unassigned work is called out, not quietly given to someone.
Prompt in Operations by AISkillHub Library
What it does
After most meetings, the notes exist but nobody is sure what was decided or who owns what. This prompt extracts three things from raw notes or a transcript: decisions, action items, and open questions. It is strict about the difference between something that was decided and something that was only discussed, and between an owner who was named and one who was assumed.
**Use it when** you have notes, a transcript or an AI meeting summary and need to send a recap or update a task tracker.
**What you get** - Decisions, each with who agreed and any conditions. - Action items in a table: action (starting with a verb), owner, due date, and the line in the notes it came from. - An "unassigned or unclear" list for tasks with no owner, proposals that were not agreed, and conflicting statements. - Open questions to carry to the next meeting. - A short recap message ready to paste into email or chat.
**What it will not do** - It will not assign an owner or due date that was not stated. It uses `[needs: owner]` and `[needs: date]` instead. Relative dates ("next week") are kept as said, with the calendar date only if the meeting date makes it unambiguous. - It will not turn a suggestion into a decision. - It will not include side comments or personal remarks in the recap.
**Example** From notes of a four-person ops sync, the prompt extracts one decision (pause a hire until January), three action items (one with no confirmed owner), and two unclear items (a holiday schedule nobody took, and a printer replacement that depends on a warranty check). See the example file.
How to use it
## How to use
1. Open `prompts/meeting-notes-to-actions.md` and copy everything below the title line. 2. Replace each `{{variable}}` with your own content. Paste long material (notes, transcripts, tables) directly in place of the variable; keep the surrounding tags so the model can tell inputs apart. 3. Start a new chat in ChatGPT, Claude or Gemini and paste the filled prompt as your first message. 4. Read the output against the quality checks at the end of the prompt before you use or send it.
## Variables
- `{{meeting_notes}}`: raw notes, a transcript, or an automated summary. - `{{meeting_date}}`: the date of the meeting, used to resolve relative dates. - `{{attendees}}`: names and roles of attendees, so owners can be matched. - `{{recap_channel}}`: `email` or `chat`. Optional; defaults to email.
If you do not have a value for an optional variable, write `none` rather than deleting it, so the prompt still reads correctly.
## Example
`prompts/meeting-notes-to-actions.example.md` shows the prompt with every variable filled and a trimmed sample of the output, so you can see the expected shape before you run it on real material.
## Reusing it
To reuse it often, save the prompt (with the variables left in) as the instructions of a Claude Project, a custom GPT or a Gemini Gem, then paste only the inputs in each new chat.
Files
Save each file at the path shown, relative to your project root.