Collection · 11 use cases
From OpenAI's own examples
The jobs OpenAI uses to explain dots, as paste-ready prompts. Official prompts are word for word; the rest are adapted and labeled.
The 11 prompts in this collection
-
Keep an offsite on track
Tracks offsite decisions, deadlines and open questions, and drafts replies for you to send.
-
Weekday planning check-in
Checks a planning channel each weekday and pings you only when something needs you.
-
Compare venues on a budget
Compares three venues on capacity, travel, dates and full price, under your cap. Books nothing.
-
Keep a sales proposal current
Keeps a proposal in step with the latest emails and call notes, showing every change.
-
Content kit from an interview
Turns one interview into exact quotes, a summary, post drafts and claims to fact-check.
-
Keep launch materials aligned
Checks landing copy, FAQ and sales material against the spec and lists every mismatch.
-
Update a study as results arrive
Updates a copy of your study as results land and notes what changed for each finding.
-
Fix reported bugs for review
Reproduces each bug report in Codex and opens a draft pull request for your review.
-
Watch Slack for bug reports
Investigates each new bug report in a Slack channel and prepares a fix, without posting.
-
Catch forgotten invoices
Matches delivered work against sent invoices and drafts the missing ones for your approval.
-
Route updates by channel
Routine progress stays in ChatGPT; decisions and risks reach you in Slack.
What’s in this collection
OpenAI’s announcement and learn docs describe what dots are for through a handful of concrete jobs. We collected the ones that make a complete, repeatable job in one place. Where OpenAI published a full prompt, it’s here word for word. Where it described the job in a sentence, we wrote the prompt and marked it as adapted.
Why start here
These are the jobs OpenAI chose to show, so they line up with what dots are designed to do today: keep a responsibility current, follow a source, and bring decisions back to you. Every entry names the official page it came from, so you can read OpenAI’s own description before you paste.
Two official prompts first
The offsite prompt and the weekday check-in are copied exactly from OpenAI’s learn docs. Try those before the adapted ones: they show the pattern OpenAI recommends, with a clear scope, drafts before sends, and a request to confirm any schedule.
Set them up: Custom rules · Connect apps · Scheduled tasks · Local computer access
From OpenAI's own examples: FAQ
Are these OpenAI's exact prompts?
Two are: the offsite prompt and the weekday check-in are copied word for word. The rest are jobs OpenAI described in a sentence, which we wrote up and labeled as adapted.
Where did OpenAI publish these examples?
In its dots announcement and the learn.chatgpt.com docs. Each use case links to the page it came from.
Which one should I try first?
The two official prompts. They show the pattern OpenAI recommends: clear scope, drafts before sends, and a confirmed schedule.