Put your recurring prompt on a schedule.
One of the easiest ways to make AI more impactful is to have routine prompts or use cases run on a schedule. That might be a regular briefing on company or industry news, a summary and tidy-up of emails or other messages, or a reminder to return to a big strategy question that keeps getting pushed aside.
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
- Suggest recurring uses from your context and verify scheduling and delivery options.
- Write, configure and test the job where supported.
What to have ready
- A recurring need, or enough context to suggest one.
Where you come in
- Choose the task, cadence and notification channel; confirm activation and delivery.
What you'll get
A configured recurring task with its next run and controls, or an explicit setup/reminder fallback.
You're done when
A recurring task is set up with the agreed cadence and delivery method, and you know how to change or stop it. Any untested first run or reminder-only fallback is explicit.
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. ### Find a recurring task worth doing I want to create a routine prompt or AI use case that runs on a schedule. Help me work out what would be useful and how to make it happen with the tools available to me. Build on any task I already have in mind. If I am unsure, say, “Let me suggest a few options.” Use relevant accessible email, chats, calendar commitments, project documents and earlier requests to identify recurring information needs, missed follow-ups or important strategy questions that lose out to urgent work. Inspect access before relying on it. Keep this search focused and distinguish what the records show from your suggestions. If access is missing, offer useful examples and ask one short question about my work. Possibilities include a company or industry news briefing, a digest of messages needing attention, an agreed inbox tidy-up, a project update, or a recurring invitation to think through a larger question. Recommend a few that fit me, then let me choose. ### Check how scheduling and delivery actually work Inspect your actual scheduling capabilities, connected apps and relevant company guidance. Search current official documentation for my AI product and account type rather than assuming that the feature exists or that it has the same capabilities as an ordinary chat. Check available models or tools, access to live sources at run time, execution limits, and whether my device must remain on or signed in. If native scheduling is unavailable or unsuitable, investigate a permitted connected automation service or calendar/device route. Distinguish a reminder that asks me to run a prompt from a workflow that really runs it unattended. [Gemini scheduled actions](https://support.google.com/gemini/answer/16316416) and [Power Automate scheduled flows](https://learn.microsoft.com/en-us/power-automate/run-scheduled-tasks) illustrate routes to investigate, not promises about my account. Verify current support and any required subscription, connection or permission. If browsing is unavailable, identify the precise information needed to verify the route. Find out how the result can reach me: email, a message, an in-app result, a notification or a popup. Verify the actual channel, destination and notification settings. Do not confuse an email saying a task finished with an email containing the result. Ask how I want to hear from it and what cadence I want. Settle the time zone, time, days, start date and any end date or quiet periods. ### Define and test the job Write a self-contained scheduled prompt with its purpose, sources to inspect each time, date window, output format, delivery route and behavior when nothing relevant happened or access failed. A future run may not have this chat's context. Make the requested output small enough to be useful at that cadence. For an ongoing digest, establish whether repeated items should be included and how the workflow will know what was already reported; do not assume it retains state. Separate reading and summarizing messages from changing them. For an inbox tidy-up, agree on exact actions and exclusions before scheduling anything that archives, marks, moves or deletes messages. Sending messages to other people also needs explicit instruction. Before creating or activating anything, check for an existing equivalent scheduled job. If one exists, show it to me and agree whether to keep or update it rather than create a duplicate. Show me the complete job and schedule, resolve the choices, then create or update it using the available supported tool once I authorize activation. If a manual setup step is necessary, guide me through that exact step. Run a small test using representative inputs. Verify the result and delivery where you can; ask me to confirm receipt where only I can see it. If only a manual run can be tested, say that the first scheduled run remains unverified. ### Confirm the recurring task is real Read back the created job or scheduler confirmation, its next run in the agreed time zone, where results will appear, and how to pause, edit or stop it. If setup fails or a required capability is absent, preserve the complete job specification and give the exact remaining action; do not claim that a recurring task exists because its prompt has been written. After the first useful result, help refine the sources, length or cadence if I want. Finish with a working scheduled task and a clear account of what has been tested, or an explicitly labeled reminder/setup fallback I have chosen. 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
Put your recurring prompt on a schedule.
One of the easiest ways to make AI more impactful is to have routine prompts or use cases run on a schedule. That might be a regular briefing on company or industry news, a summary and tidy-up of emails or other messages, or a reminder to return to a big strategy question that keeps getting pushed aside.
How this works depends on the tool you use. Some can run scheduled work themselves; others need a connected automation service. A calendar reminder can bring you back to a task, though it does not necessarily run the AI for you. The prompt below will help you find out what is available and choose something useful.
For the user
Start with a recurring task you have in mind, or ask the AI to suggest a few. You will choose the task, cadence and way you want to hear from it.
Done when
A recurring task is set up with the agreed cadence and delivery method, and you know how to change or stop it. Any untested first run or reminder-only fallback is explicit.
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
Find a recurring task worth doing
I want to create a routine prompt or AI use case that runs on a schedule. Help me work out what would be useful and how to make it happen with the tools available to me.
Build on any task I already have in mind. If I am unsure, say, “Let me suggest a few options.” Use relevant accessible email, chats, calendar commitments, project documents and earlier requests to identify recurring information needs, missed follow-ups or important strategy questions that lose out to urgent work. Inspect access before relying on it. Keep this search focused and distinguish what the records show from your suggestions. If access is missing, offer useful examples and ask one short question about my work.
Possibilities include a company or industry news briefing, a digest of messages needing attention, an agreed inbox tidy-up, a project update, or a recurring invitation to think through a larger question. Recommend a few that fit me, then let me choose.
Check how scheduling and delivery actually work
Inspect your actual scheduling capabilities, connected apps and relevant company guidance. Search current official documentation for my AI product and account type rather than assuming that the feature exists or that it has the same capabilities as an ordinary chat. Check available models or tools, access to live sources at run time, execution limits, and whether my device must remain on or signed in.
If native scheduling is unavailable or unsuitable, investigate a permitted connected automation service or calendar/device route. Distinguish a reminder that asks me to run a prompt from a workflow that really runs it unattended. Gemini scheduled actions and Power Automate scheduled flows illustrate routes to investigate, not promises about my account. Verify current support and any required subscription, connection or permission. If browsing is unavailable, identify the precise information needed to verify the route.
Find out how the result can reach me: email, a message, an in-app result, a notification or a popup. Verify the actual channel, destination and notification settings. Do not confuse an email saying a task finished with an email containing the result. Ask how I want to hear from it and what cadence I want. Settle the time zone, time, days, start date and any end date or quiet periods.
Define and test the job
Write a self-contained scheduled prompt with its purpose, sources to inspect each time, date window, output format, delivery route and behavior when nothing relevant happened or access failed. A future run may not have this chat's context. Make the requested output small enough to be useful at that cadence. For an ongoing digest, establish whether repeated items should be included and how the workflow will know what was already reported; do not assume it retains state.
Separate reading and summarizing messages from changing them. For an inbox tidy-up, agree on exact actions and exclusions before scheduling anything that archives, marks, moves or deletes messages. Sending messages to other people also needs explicit instruction.
Before creating or activating anything, check for an existing equivalent scheduled job. If one exists, show it to me and agree whether to keep or update it rather than create a duplicate. Show me the complete job and schedule, resolve the choices, then create or update it using the available supported tool once I authorize activation. If a manual setup step is necessary, guide me through that exact step. Run a small test using representative inputs. Verify the result and delivery where you can; ask me to confirm receipt where only I can see it. If only a manual run can be tested, say that the first scheduled run remains unverified.
Confirm the recurring task is real
Read back the created job or scheduler confirmation, its next run in the agreed time zone, where results will appear, and how to pause, edit or stop it. If setup fails or a required capability is absent, preserve the complete job specification and give the exact remaining action; do not claim that a recurring task exists because its prompt has been written.
After the first useful result, help refine the sources, length or cadence if I want. Finish with a working scheduled task and a clear account of what has been tested, or an explicitly labeled reminder/setup fallback I have chosen.
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.