Last updated:
Quick answer
List your repetitive weekly tasks, score each for hours spent, how often it repeats and how bad a mistake would be, and start with the one that is frequent, slow and low risk, usually drafting replies or sorting the inbox. Run it with approval on for 30 days before adding a second job.
Free PDF
Get this setup as a PDF
Confirm your email and we'll open a print-ready copy of this setup you can save as a PDF. You also get the free starter pack and one setup a week. Unsubscribe anytime.
Already a subscriber? Unlock it on this device
At a glance
All outcome figures are estimates from cited sources, not independent verification by us.
- What it is allowed to do
- Read-only
Only reads, researches, or summarizes. It never drafts outbound content or changes anything.
- Time saved (estimate)
- About 30 minutes to plan.
- Difficulty
- Beginner
- Tools needed
- ChatGPT, Claude
- Typical cost
- Free tiers can plan it. Agents that act may need a paid plan; prices: Not stated in sources.
Reviewed Nov 2, 2026
Basis (estimate): Estimate only: based on the steps above, not timed.
What does this setup do?
Score your weekly tasks for time, repetition and risk, and pick one safe first job for an AI agent, with a 30-day test plan and a clear way to measure if it helped.
Who is it for?
Owners who keep hearing about AI agents but do not know which job to hand over first.
How do you set it up?
Tools
- ChatGPT or Claude
- A list of your weekly tasks
Steps
- Write down every task you do more than once a week, with rough minutes each time.
- Paste the list into the prompt below. It scores each task for time saved, repetition and risk.
- Pick the top task that is low risk. Drafting, sorting and summarizing are safer first jobs than sending, paying or deleting.
- Before connecting anything, remember OpenAI says a signed-in agent can access sensitive data and perform actions. Keep approval on for anything that leaves the business.
- Run it for 30 days. Track minutes saved and every mistake.
- Add a second job only after the first one runs a month without surprises.
Example output
Illustrative. Your results will differ.
- 1. Sort and draft replies to the inbox: 5+5-2 = 8
- 2. Weekly status update: 3+4-1 = 6
- 3. Pay vendor bills: 4+3-5 = 2 (not first)
- First job: inbox drafts, approve before send, measure minutes saved weekly
Where to paste this prompt
ChatGPT → Project instructions, or ChatGPT Tasks for scheduled reminders (once a day on Free)
Copy-paste prompt
Adapt bracketed fields to your situation.
Help me pick the first job to hand to an AI agent.
My weekly tasks (task, minutes each time, times per week): [PASTE LIST]
For each task, score 1 to 5:
- Time it eats per week
- How repetitive it is
- Risk if the AI gets it wrong (5 = very risky)
Then:
1. Rank the tasks by (time + repetition) minus risk.
2. Recommend one first job and say why.
3. Write a 30-day test plan: what the agent does, what I approve, and what I measure each week.
Use only my list. Do not suggest tools I did not mention.Newsletter
Get the free starter pack
Free starter pack with five beginner copy-paste prompts after you confirm, plus one researched setup in your inbox each week. Unsubscribe anytime.
What usually goes wrong?
Where agents tend to stall on this job. More patterns in Where AI gets stuck.
- Starting with something risky, like payments, because it saves the most time.
- Connecting every account at once.
- Skipping the measuring, then not knowing if it helped.
Which chat tools work with this prompt?
Paste the prompt into any of these assistants (free or paid tiers).
- ChatGPT
- Claude
Common questions
- How do I start using AI agents in my business?
- Pick one frequent, low-risk task, like drafting replies, run it with approval on for 30 days, and measure time saved before adding more.
- What is a good first job for an AI agent?
- Drafting, sorting and summarizing. Leave sending, paying and deleting behind your approval.
- Do I need to code?
- No. Many assistants connect to email and calendar from their settings.
What tips make it smoother?
- The best first job is boring, frequent and easy to check.
- Measure minutes saved; it keeps the decision honest.
- One job running well beats five half-set-up agents.
Where do the claims come from?
Summaries use public sources linked below, not independent verification. How setups are picked.
- Read OpenAI’s write-up (help.openai.com)
OpenAI: ChatGPT agent: when signed into sites or apps it can access sensitive data like emails, files or account settings and perform actions (read 2026-11-02 from a search result excerpt of the page)
- Read Anthropic’s write-up (support.claude.com)
Anthropic: Use Google Workspace connectors: Gmail drafting, approval before send, reply or forward by default (read 2026-10-31 through a page fetch tool, not a direct browser open)
Related setups
- Set guardrails before an AI agent gets access
Write a one-page rulebook for an AI agent before it touches your accounts: what it can read, what it can only draft, what always needs your approval, and what it must never do.
- Make your AI agent stop for approval
Make your AI agent stop and show you every email, booking or post as a draft with a one-line summary, and wait for your OK before it goes out.
- Morning inbox briefing and multi-inbox triage
Start the day with a short “what needs you” note, and periodically sort multiple inboxes into reply vs FYI vs junk, with drafts for your approval, not auto-send.