Have meeting follow-ups write themselves.
A meeting can sound decisive while leaving the work surprisingly unclear. Turn what people actually agreed into a follow-up that makes the next steps easy to act on.
Try it now
Copy the prompt into your AI tool of choice, then answer its questions as they come.
Paste it into any AI chat to start. For most people, this is all you need.
Everything on this page in one paste: the prompt plus the full plan, every step and what to check. Use it for longer moves, to plan ahead, to share with a colleague, or if your AI can't open links.
How it goes
What the AI will do
- Read the meeting record and project context to distinguish decisions, suggestions and commitments.
- Write a ready-to-send follow-up and tailored reminders for the people with actions.
What to have ready
- A meeting name, invitation or transcript; the AI will look for linked notes, documents and task records.
- Transcript or notes access, plus relevant correspondence, makes owners and deadlines easier to verify.
Where you come in
- Settle consequential gaps such as an unclear owner, date or recipient.
- Review the drafts and authorize sending if you want the AI to send them and it has that capability.
What you'll get
A checked meeting follow-up and individual action reminders, ready to send—or sent only after your authorization—with remaining uncertainties identified.
You're done when
You have a checked follow-up and personalized reminders that reflect the meeting evidence, with owners, dates and uncertainties clear; any sending is separately authorized and verified.
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result. ### Start with the activity and the user's context Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear. Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them. ### Check the capabilities needed for this work Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit. When useful access is missing: - Explain what you cannot reach and how that material would help. - Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot. - If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response. - Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available. ### Ask focused questions and respect the answers Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful. Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements. ### Keep evidence and independent contributions distinct Distinguish evidence, inference, and missing information. Check your work against the source material. Use independent agents when your system actually supports delegation and independent investigation or review would improve the work: - Give each agent a bounded brief and the context appropriate to its role. - Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations. - Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately. ### Produce, check, and revise the result Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them. Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result. Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized. Help me turn a meeting into an accurate follow-up and individual action reminders. Do the investigation and drafting, keep proposals distinct from commitments, and ask questions one at a time when a missing answer matters. ### 1. Find the meeting and its purpose Start with the meeting name, date, transcript or link I supplied. Inspect available calendar, transcript, document, email, chat and task tools using small relevant reads. Locate the invitation and agenda to establish purpose, participants and the intended decision. Search the linked project documents, preceding emails and chats, including relevant WhatsApp messages where accessible, for the assignment, terminology, constraints and earlier commitments. Search subsequent messages and task updates for clarifications that already resolve apparent gaps. Use those findings to establish what the follow-up should accomplish and who needs it. Distinguish invitees from actual attendees when the record allows; identify an absent owner who still needs an action reminder. Ask me to resolve an uncertain meeting identity or consequential audience choice before drafting for that audience. Retrieve the transcript and check its coverage, speaker labels and relation to any recording or official minutes. If unavailable, inspect meeting notes and subsequent participant accounts. Investigate accessible export or connection routes before asking me for material. Explain that notes alone may not establish precise wording, agreement or attribution. If only my recollection is available, build a clearly labeled draft from it and ask me to confirm commitments represented as agreed. ### 2. Extract the meeting's actual business Read the full relevant discussion, including context around proposed actions and the close of the meeting. Create a working record with source references for: - Decisions, the question each answers, conditions and effective dates. - Explicit commitments, each owner, deliverable and deadline. - Suggestions, requests and unresolved questions that were never accepted. - Dependencies or decisions someone needs before they can act. Do not promote “we should” into an assigned action. Check pronouns and speaker attribution against surrounding exchanges. Preserve a stated date's timezone where relevant, and resolve relative dates against the meeting date. If a transcript says “next Friday” ambiguously, use subsequent confirmation or ask rather than silently choosing. Compare the extraction with official minutes, project documents and later messages. Record actual amendments with their sources; a later file timestamp alone does not establish a new agreement. Link claims to transcript timestamps, document sections or messages so the drafts can be checked. For a long meeting or several meetings, split inspection by bounded agenda item or meeting. Give each agent the actual relevant records, extraction requirements and enough surrounding context. Track topic and source coverage, then reconcile duplicate or conflicting actions centrally. Use waves as needed. With no agents, use manageable passes or fresh-session handoffs and disclose when checking shares the drafting context. ### 3. Resolve the few gaps that affect action Show me consequential ambiguities plainly: for example, one person offered help but no owner accepted, or two records give different due dates. Ask one focused question at a time and wait before drafting that item as settled. Recommend wording such as “Owner to confirm” when a useful follow-up can surface the gap without inventing a decision; establish with me whether that is the desired purpose. Use existing answers and approvals rather than repeatedly seeking confirmation. Continue drafting uncontested material while waiting. If no transcript or reliable notes exist, obtain my confirmation of the consequential recollected decisions and commitments before representing them as the meeting's conclusions. ### 4. Write messages people can act on Produce the actual meeting follow-up in a suitable email or chat format. Lead with the outcome, give only the context recipients need, and make decisions and next steps easy to scan. Include owners, deliverables, agreed dates and unresolved asks. Use the team's terminology and an appropriate tone inferred from relevant correspondence; keep background detail behind source links when possible. Create an individual reminder for each person with an action. Give that person their deliverable, date, necessary dependency and where to find supporting material. Combine related actions for the same recipient rather than generating a message for every line. Mark a proposed reminder schedule as a suggestion unless already agreed. Verify recipient identities and destinations from actual records; do not infer an address from a name. Keep the complete follow-up and individual reminders consistent. A tailored reminder must not change the obligation, add a deadline or imply an agreement missing from the common record. ### 5. Review, deliver and handle sending Give an independent reviewer the actual drafts and meeting evidence, without your drafting rationale. Ask it to verify decisions, attribution, dates, recipients, dependencies and any difference between the follow-up and reminders. For a large set, review in bounded groups while retaining the common action record. Use a fresh-session fallback if possible, or identify the limits of a sequential same-context review. Repair material errors and inspect the revised messages. Present the finished drafts with a concise note about any remaining uncertainty. Ask for my review of consequential choices and authorization before sending, unless I have already explicitly authorized these exact messages and recipients. If I authorize sending and your tools support it, send the approved messages and inspect the result; report what was sent and any failures. Otherwise leave ready-to-send drafts with verified recipients. Do not describe drafts as sent or reminders as scheduled unless those actions actually succeeded. These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
View full AI instructions
Have meeting follow-ups write themselves.
What the AI will do
- Read the meeting record and project context to distinguish decisions, suggestions and commitments.
- Write a ready-to-send follow-up and tailored reminders for the people with actions.
What you can provide
- A meeting name, invitation or transcript; the AI will look for linked notes, documents and task records.
- Transcript or notes access, plus relevant correspondence, makes owners and deadlines easier to verify.
Where you'll take part
- Settle consequential gaps such as an unclear owner, date or recipient.
- Review the drafts and authorize sending if you want the AI to send them and it has that capability.
What you'll get
A checked meeting follow-up and individual action reminders, ready to send—or sent only after your authorization—with remaining uncertainties identified.
Activity context
A meeting can sound decisive while leaving the work surprisingly unclear. Turn what people actually agreed into a follow-up that makes the next steps easy to act on.
For the user
Identify the meeting or provide whatever record you have. The AI will find related context, draft the follow-up and flag missing decisions. Expect to settle only the gaps that matter, then review the complete messages before any sending.
Guidance for the AI
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.
Start with the activity and the user's context
Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.
Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.
Check the capabilities needed for this work
Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.
When useful access is missing:
- Explain what you cannot reach and how that material would help.
- Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
- If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
- Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.
Ask focused questions and respect the answers
Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.
Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.
Keep evidence and independent contributions distinct
Distinguish evidence, inference, and missing information. Check your work against the source material.
Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:
- Give each agent a bounded brief and the context appropriate to its role.
- Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
- Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.
Produce, check, and revise the result
Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.
Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.
Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.
Your AI prompt
Help me turn a meeting into an accurate follow-up and individual action reminders. Do the investigation and drafting, keep proposals distinct from commitments, and ask questions one at a time when a missing answer matters.
1. Find the meeting and its purpose
Start with the meeting name, date, transcript or link I supplied. Inspect available calendar, transcript, document, email, chat and task tools using small relevant reads. Locate the invitation and agenda to establish purpose, participants and the intended decision. Search the linked project documents, preceding emails and chats, including relevant WhatsApp messages where accessible, for the assignment, terminology, constraints and earlier commitments. Search subsequent messages and task updates for clarifications that already resolve apparent gaps.
Use those findings to establish what the follow-up should accomplish and who needs it. Distinguish invitees from actual attendees when the record allows; identify an absent owner who still needs an action reminder. Ask me to resolve an uncertain meeting identity or consequential audience choice before drafting for that audience.
Retrieve the transcript and check its coverage, speaker labels and relation to any recording or official minutes. If unavailable, inspect meeting notes and subsequent participant accounts. Investigate accessible export or connection routes before asking me for material. Explain that notes alone may not establish precise wording, agreement or attribution. If only my recollection is available, build a clearly labeled draft from it and ask me to confirm commitments represented as agreed.
2. Extract the meeting's actual business
Read the full relevant discussion, including context around proposed actions and the close of the meeting. Create a working record with source references for:
- Decisions, the question each answers, conditions and effective dates.
- Explicit commitments, each owner, deliverable and deadline.
- Suggestions, requests and unresolved questions that were never accepted.
- Dependencies or decisions someone needs before they can act.
Do not promote “we should” into an assigned action. Check pronouns and speaker attribution against surrounding exchanges. Preserve a stated date's timezone where relevant, and resolve relative dates against the meeting date. If a transcript says “next Friday” ambiguously, use subsequent confirmation or ask rather than silently choosing.
Compare the extraction with official minutes, project documents and later messages. Record actual amendments with their sources; a later file timestamp alone does not establish a new agreement. Link claims to transcript timestamps, document sections or messages so the drafts can be checked.
For a long meeting or several meetings, split inspection by bounded agenda item or meeting. Give each agent the actual relevant records, extraction requirements and enough surrounding context. Track topic and source coverage, then reconcile duplicate or conflicting actions centrally. Use waves as needed. With no agents, use manageable passes or fresh-session handoffs and disclose when checking shares the drafting context.
3. Resolve the few gaps that affect action
Show me consequential ambiguities plainly: for example, one person offered help but no owner accepted, or two records give different due dates. Ask one focused question at a time and wait before drafting that item as settled. Recommend wording such as “Owner to confirm” when a useful follow-up can surface the gap without inventing a decision; establish with me whether that is the desired purpose.
Use existing answers and approvals rather than repeatedly seeking confirmation. Continue drafting uncontested material while waiting. If no transcript or reliable notes exist, obtain my confirmation of the consequential recollected decisions and commitments before representing them as the meeting's conclusions.
4. Write messages people can act on
Produce the actual meeting follow-up in a suitable email or chat format. Lead with the outcome, give only the context recipients need, and make decisions and next steps easy to scan. Include owners, deliverables, agreed dates and unresolved asks. Use the team's terminology and an appropriate tone inferred from relevant correspondence; keep background detail behind source links when possible.
Create an individual reminder for each person with an action. Give that person their deliverable, date, necessary dependency and where to find supporting material. Combine related actions for the same recipient rather than generating a message for every line. Mark a proposed reminder schedule as a suggestion unless already agreed. Verify recipient identities and destinations from actual records; do not infer an address from a name.
Keep the complete follow-up and individual reminders consistent. A tailored reminder must not change the obligation, add a deadline or imply an agreement missing from the common record.
5. Review, deliver and handle sending
Give an independent reviewer the actual drafts and meeting evidence, without your drafting rationale. Ask it to verify decisions, attribution, dates, recipients, dependencies and any difference between the follow-up and reminders. For a large set, review in bounded groups while retaining the common action record. Use a fresh-session fallback if possible, or identify the limits of a sequential same-context review. Repair material errors and inspect the revised messages.
Present the finished drafts with a concise note about any remaining uncertainty. Ask for my review of consequential choices and authorization before sending, unless I have already explicitly authorized these exact messages and recipients. If I authorize sending and your tools support it, send the approved messages and inspect the result; report what was sent and any failures. Otherwise leave ready-to-send drafts with verified recipients. Do not describe drafts as sent or reminders as scheduled unless those actions actually succeeded.
Done when
You have a checked follow-up and personalized reminders that reflect the meeting evidence, with owners, dates and uncertainties clear; any sending is separately authorized and verified.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.