DISCOVER ON TOOLFINDINGSAirbnbFreeTravel & Ride Apps profile covering workflow fit, platforms, pricing model, strengths and limitations.Explore Tool
Operations · Intermediate

Meeting Decision Extractor

Meeting Decision Extractor is a practical, reusable AI prompt for meetings, operations, verification. For “Meeting Decision Extractor”, apply this guidance to the exact input, audience and output required: This expanded guide turns the prompt page into a complete working reference: what to prepare, how to use the prompt, how to improve the first response and how to validate the final result.

When to use this prompt

For “Meeting Decision Extractor”, use this task-specific prompting checkpoint: use it when you need a repeatable result with clear constraints instead of a vague one-line instruction. For “Meeting Decision Extractor”, apply this guidance to the exact input, audience and output required: Before starting, define the business or personal objective, the intended audience, relevant background information and the exact deliverable you expect.

Information to provide

  • Objective: state the outcome, decision or asset you need.
  • Context: When using “Meeting Decision Extractor”, apply this prompt-specific requirement: provide facts, source material and background the model must respect.
  • Audience: identify who will read, approve or act on the result.
  • Constraints: For “Meeting Decision Extractor”, use this task-specific prompting checkpoint: specify tone, scope, length, exclusions, policies or brand requirements.
  • Output structure: To improve “Meeting Decision Extractor”, define this point for the actual input and output: request the exact headings, steps, fields, table columns or format you need.
  • Acceptance criteria: describe what a useful final answer must include.

Step-by-step usage

To improve “Meeting Decision Extractor”, define this point for the actual input and output: replace every placeholder with concrete information before submitting the prompt. To improve “Meeting Decision Extractor”, connect this principle to the requested format and downstream use: Run the first version, then compare the response against the acceptance criteria rather than judging it only by fluency. For “Meeting Decision Extractor”, apply this guidance to the exact input, audience and output required: If something is missing, make a targeted follow-up request that identifies the exact section and correction required.

Refinement strategy

When using “Meeting Decision Extractor”, apply this prompt-specific requirement: for style-sensitive work, provide a short example of the desired voice or format. When using “Meeting Decision Extractor”, make this checkpoint specific to the source material and acceptance criteria: For fact-sensitive work, supply authoritative source material and ask the model to distinguish supplied facts from assumptions or recommendations. For “Meeting Decision Extractor”, interpret this guidance in the context of the real task rather than as a generic prompting rule: Ask for alternatives only when they help the decision; otherwise keep the model focused on improving the selected direction.

Model and difficulty guidance

This prompt is classified as intermediate. Listed model guidance: ChatGPT,Claude,Gemini. Related topics include meetings, operations, verification. When using “Meeting Decision Extractor”, make this checkpoint specific to the source material and acceptance criteria: Different assistants may interpret the same wording differently, so preserve the objective, inputs, constraints and output specification when adapting it.

Accuracy and safety checks

AI output can contain outdated or invented details. For “Meeting Decision Extractor”, interpret this guidance in the context of the real task rather than as a generic prompting rule: Independently verify important names, dates, figures, quotations, links and claims. For “Meeting Decision Extractor”, apply this guidance to the exact input, audience and output required: For legal, medical, financial, security or other high-impact decisions, use appropriate qualified review rather than relying on generated text alone.

Final checklist

  • All placeholders contain real information.
  • When using “Meeting Decision Extractor”, apply this prompt-specific requirement: the response addresses the stated objective and audience.
  • The requested structure and constraints were followed.
  • Important factual claims were checked.
  • Unsupported assumptions were corrected or removed.
  • When using “Meeting Decision Extractor”, apply this prompt-specific requirement: the final output was edited for clarity, relevance and practical use.
ChatGPT Claude Gemini
Full prompt

Prompt text

You are an operations analyst responsible for turning meeting transcripts into a precise, evidence-backed decision and action register.

MEETING TRANSCRIPT
{transcript}

OBJECTIVE
Analyse the transcript and identify only information supported by the conversation. Do not convert suggestions, assumptions or unresolved discussion into confirmed decisions.

1. CONFIRMED DECISIONS
Extract every decision that was clearly agreed, approved or finalised.

For each provide:
- decision
- decision owner if stated
- supporting transcript evidence
- confidence: Confirmed / Needs confirmation

Do not classify an idea or recommendation as a decision unless the transcript clearly shows agreement.

2. ACTION ITEMS
Extract every concrete task or follow-up commitment.

For each provide:
- action
- owner
- deadline
- priority if evident
- dependencies if mentioned
- supporting transcript evidence
- status

If an action is clear but no owner or deadline was assigned, write:
Owner: Not assigned
Deadline: Not specified

Never invent an owner or deadline.

3. OPEN QUESTIONS
List unresolved questions, decisions still pending and issues requiring additional information.

For each provide:
- question
- why unresolved
- responsible person if stated
- supporting evidence

4. RISKS AND BLOCKERS
Identify blockers, dependencies, disagreements or risks that could prevent agreed work from moving forward.

Separate confirmed blockers from possible concerns.

5. FOLLOW-UP REQUIRED
Identify:
- unclear ownership
- missing deadlines
- contradictory statements
- provisional decisions
- promised information not yet provided

6. EVIDENCE RULE
Every important extracted item must include a short supporting transcript excerpt.

Do not invent quotations.

7. ACCURACY RULES
- Do not infer agreement because nobody objected.
- Do not turn brainstorming into commitments.
- Preserve names, dates, numbers and conditions.
- If statements conflict, report the conflict.
- Distinguish facts, decisions, actions and suggestions.

OUTPUT FORMAT

## Executive Summary
3–6 concise bullets.

## Confirmed Decisions
| # | Decision | Owner | Evidence | Confidence |

## Action Items
| # | Action | Owner | Deadline | Dependencies | Evidence | Status |

## Open Questions
| # | Question | Why unresolved | Responsible person | Evidence |

## Risks & Blockers
| # | Risk / Blocker | Impact | Evidence | Status |

## Follow-Up Required

## Final Verification
Confirmed decisions: [number]
Action items: [number]
Open questions: [number]
Items needing confirmation: [number]
Expected structure

What a useful answer should look like

An evidence-backed meeting register containing executive summary, confirmed decisions, action items, owners, deadlines, open questions, blockers, supporting transcript evidence and a final verification count.
Use responsibly

How to get a better result

Add real context

Replace every placeholder and include constraints, audience and source material.

Verify claims

Open original sources for factual, legal, medical, financial or time-sensitive information.

Iterate deliberately

Ask for revisions against a rubric instead of accepting the first output.

Questions from ToolQuestions

Practical ChatGPT Q&A

Continue your research with relevant published questions and concise answers from ToolQuestions.

View all questions ↗
Page 1 of 7
Questions from ToolQuestions

Practical Claude Q&A

Continue your research with relevant published questions and concise answers from ToolQuestions.

View all questions ↗
?
Business

What should I document after a Claude pilot?

For Claude, regarding “What should I document after a Claude pilot?”: Treat Claude as a team process, not only a tool: assign an owner, define review rules, document permissions, measure outcomes and keep a fallback f...

Read answer →
Page 1 of 8

Related research and practical resources

Move between explanation, evaluation and practical use without losing context.