Scheduled Prompts

Set up once. AI runs it every day

Prompts built for recurring execution in ChatGPT Tasks, Copilot Tasks, and Claude Cowork. Configure once and they run automatically without re-prompting.

ChatGPT TasksCopilot TasksClaude Cowork
See the twelve playsSee pricing

The execution ladder

Five rungs, from a single ask to a standing routine

A scheduled prompt is the last rung. It is the same craft as a good prompt, with a time attached and nobody left to remember it.

How it works

Three steps, once

1

Pick a play

Choose one below. Read what it produces and what it needs before you commit to a cadence.

2

Paste it into your scheduler

ChatGPT Tasks, Copilot Tasks, or Claude Cowork — each one has a place to save a prompt and give it a time.

3

Let it run, then read the output

The point is the output waiting for you, not the prompting. Change the cadence when your week changes.

New to this? Set up your assistant first, or read how Claude Cowork handles standing work.

The library

Twelve plays that run on a schedule

Each one is written to be pasted into your assistant’s task scheduler once and left alone. Cards are open to everyone; the full prompt and its setup notes come with Pro.

Daily

Runs every day

DailyCopilot Tasks

The Copilot Pre-Call Morning Ritual

A daily 10–20 minute morning routine for sellers using Microsoft Copilot to sweep the pipeline, prep for every calendar meeting, and surface open commitments — before touching email.

You getfull pipeline awareness in under 2 minutes, 15–20 minutes saved per meeting prep, rep enters every call prepared not reactive

AE · 10–20 min

ProChecking your access…
DailyClaude Cowork

Daily Meeting Prep Sweep

One morning run that briefs you on every external meeting on today's calendar.

You geta brief for every external meeting today, zero walk-in-cold moments

AE · 10 min each morning

ProChecking your access…

Weekly

Runs every week

WeeklyChatGPT TasksCopilot TasksClaude Cowork

Weekly Deal Inspection and Stalled-Deal Rescue

Inspect what actually changed each week, catch stalls early, and either rescue the deal or downgrade it honestly — seven prompts.

You getHonest forecasts, earlier stall detection, cleaner pipeline hygiene, and a defensible weekly deal review.

Sales Manager · Moderate

ProChecking your access…
WeeklyChatGPT TasksCopilot TasksClaude Cowork

MEDDPICC Weekly Deal Hygiene Ritual

A weekly AI-assisted ritual for reviewing every open deal against MEDDPICC criteria — identifying gaps, flagging stale deals, and planning targeted interventions before the weekly pipeline call.

You getevery deal reviewed against MEDDPICC weekly, stale deals flagged before they slip, qualification gaps identified and actioned

AE · 30 min weekly

ProChecking your access…
WeeklyChatGPT TasksCopilot TasksClaude Cowork

Pipeline Velocity & Health Report Workflow

A weekly workflow that turns CRM pipeline velocity data into a health report — identifying stage bottlenecks, conversion drops, and rep-level coaching opportunities.

You getpipeline velocity bottlenecks identified weekly, stage conversion rates visible, rep-level coaching focus areas identified

Sales Manager · 20 min

ProChecking your access…
WeeklyChatGPT TasksCopilot TasksClaude Cowork

Rep Self-Coaching & Feedback Loop Workflow

A weekly self-coaching workflow that helps reps identify behavioral patterns from their own sales data and define one high-impact improvement to focus on each week.

You getweekly coaching habit formed without manager dependency, behavioral patterns identified from own data, single improvement focus per week drives compound growth

AE · 20 min weekly

ProChecking your access…
WeeklyClaude Cowork

Living Objection Library

An objection library that maintains itself from your real calls — every new objection captured, every response graded by what actually happened next.

You getobjection library that grows from real calls, responses that actually worked

AE · 15 min per week

ProChecking your access…
WeeklyClaude Cowork

Deals That Coach Themselves

A weekly sweep where every active deal gets the scrutiny a great manager would give it — risks, gaps, and one next best action each.

You getweekly deal-by-deal risk review, next best action per deal, no quiet stalls

AE · 15 min per week

ProChecking your access…

Monthly

Runs every month

MonthlyClaude Cowork

ICP That Sharpens Itself

Let your closed-won and closed-lost history argue with your ICP document — and update it with evidence, monthly.

You getevidence-based ICP updates, scoring rules that match reality, better-fit pipeline

Sales Manager · 30 min per quarter-month

ProChecking your access…
MonthlyClaude Cowork

10-K and Earnings Watcher

Watch your top enterprise accounts' SEC filings and earnings calls for the risk factors and initiatives your product addresses.

You getfiling summaries mapped to your pitch, exec-ready talking points, monthly refresh

AE · 15 min per run

ProChecking your access…
MonthlyClaude Cowork

Competitive Intelligence Refresh

Monthly battlecard refresh built from your actual closed-lost deals plus live web research on each competitor.

You getcurrent battlecards from real loss data, trap-setting discovery questions, refreshed monthly

AE · 20 min per month

ProChecking your access…
MonthlyClaude Cowork

Customer Feedback to Renewal Risk

Read a month of support conversations and NPS verbatims, and walk away with a ranked at-risk list and a save play for each account.

You getat-risk account list with evidence, save-play per account, renewal surprises eliminated

CSM · 20 min per month

ProChecking your access…

Where these run

Three schedulers, three habits

ChatGPT Tasks

Tasks live in the ChatGPT model picker — choose the tasks-capable model, write the prompt, then set when it should run.

One thing to watchTasks run without you in the room, so the prompt has to name its own inputs. Say where the data comes from rather than assuming a conversation is already open.

Set up ChatGPT →

Copilot Tasks

Copilot schedules work against Microsoft 365 — Outlook, Teams, Dynamics and SharePoint are already connected once you are signed in.

One thing to watchCopilot answers from what your account can see. Scope the prompt to named accounts or a saved view, or you will get the whole tenant.

Set up Copilot →

Claude Cowork

Cowork keeps a workspace of files and connectors, so a standing task can read the same folder every week and write its output back.

One thing to watchCowork rewards a workspace kept tidy. Give the task a folder to read and a file to update, and the run becomes repeatable instead of improvised.

How Cowork works →

Questions

Before you schedule anything

What is a scheduled prompt?

A prompt written to run on a cadence inside an AI assistant's own task scheduler, rather than one you type when you happen to think of it. It names its own inputs and produces the same shape of output every time, so the result is waiting for you instead of being something you have to go and ask for.

How is this different from an automated workflow?

An automated workflow is run by an external tool such as Zapier, Make or Power Automate, and usually moves data between systems. A scheduled prompt is run by the assistant itself. If you can paste it into ChatGPT Tasks, Copilot Tasks or Claude Cowork and give it a time, it belongs here. If it needs a connector platform to fire, it belongs on automated workflows.

ChatGPT Tasks, Copilot Tasks or Claude Cowork — which should I use?

Use whichever one already sees your data. Copilot is the shortest path if your pipeline lives in Microsoft 365. Claude Cowork suits standing work that reads and writes the same files each week. ChatGPT Tasks is the most portable when the inputs are ones you paste in yourself.

Do I need Pro to use these?

Yes. Every scheduled prompt on this page is part of Pro. The cards, cadences and outcomes stay open to everyone so you can see exactly what you would be getting; the prompt itself and its setup notes are for Pro members. Pro is $12 per month or $89 per year; lifetime access is listed on the pricing page.

How do I set one up?

Copy the prompt, open your assistant's scheduler, paste it in and give it a time. Run it manually once first: if the output is not what you want, the prompt is easier to fix before it is on a schedule than after.

What if the output drifts over time?

Re-read the output monthly. Scheduled prompts fail quietly rather than loudly — the run keeps succeeding while the answer slowly stops being useful. Most drift is fixed by naming the source more tightly, not by making the prompt longer.

Keep going

The other four rungs

Browse the library →

Every recurring play, in full

Pro opens the full prompt and setup notes on every scheduled play here — and the rest of the library with it: the prompts, the chains, the workflows, and the automations.

$12 per month or $89 per year.