Inbox, calendar & follow-ups
Morning brief from inbox and calendar
Every weekday before work, your dot reads today's calendar and overnight email and sends one short brief: meetings, prep, replies owed.
Copy a prompt, paste it into your dot, and it gets to work. 31 use cases, each labeled with where it came from.
Inbox, calendar & follow-ups
Every weekday before work, your dot reads today's calendar and overnight email and sends one short brief: meetings, prep, replies owed.
Projects & planning
Hand your dot the plan and venue emails; it tracks decisions, deadlines and open questions, and drafts replies for you to send.
Inbox, calendar & follow-ups
Twice a day your dot sorts new mail into needs me, FYI and ignore, and drafts replies for the needs-me pile. It never sends or archives.
Research & monitoring
Once a week your dot checks named public pricing pages in its cloud browser and reports only what changed, with before and after.
Developers
Your dot works through bug reports, reproduces each one in Codex, and opens a draft pull request with the fix for you to review.
Personal & home
Once a day your dot checks the product pages you list and messages you only when a price falls below your target. It never buys.
11 from OpenAI's own examples · 2 official prompts word for word · checked Oct 1, 2026 See all 31 →
The jobs OpenAI uses to explain dots, turned into prompts you can paste. Official prompts are word for word; the rest are marked as adapted.
Hand your dot the plan and venue emails; it tracks decisions, deadlines and open questions, and drafts replies for you to send.
Your dot matches delivered work against invoices you've sent and drafts the missing ones. Nothing goes to a client until you approve it.
Routine progress stays in ChatGPT; decisions, risks and approvals reach you in Slack with the options and a deadline.
Your dot compares three venues on capacity, travel, dates and full price, keeps the total under your cap, and books nothing.
Each weekday morning your dot reads a planning channel, updates the list, and messages you only when a deadline or decision needs you.
When new results land, your dot updates a copy of the study, reruns the summary numbers, and notes what changed for each finding.
Your dot keeps a working copy of a proposal in step with the latest emails and call notes, and shows you every change and its source.
Share an interview and your dot returns a kit: exact quotes, a summary, post drafts, a newsletter blurb and claims to fact-check.
Your dot checks landing copy, FAQ and sales material against the product spec and lists every mismatch with a suggested fix.
Your dot works through bug reports, reproduces each one in Codex, and opens a draft pull request with the fix for you to review.
When a new bug report lands in a Slack channel, your dot investigates it and prepares a fix for your review, without posting in the channel.
Sets of jobs that belong together. Good if you don't know where to start.
Five low-risk jobs to learn how your dot works before you trust it with anything that reaches other people.
The jobs OpenAI uses to explain dots, as paste-ready prompts. Official prompts are word for word; the rest are adapted and labeled.
A chief-of-staff starter set: brief, triage, prep, nudges and invoices. Everything that would reach another person stays a draft.
Jobs where your dot only looks and reports: prices, competitors, mentions, pull requests, traffic. A safe way to try scheduled work.
Feedback in, reviewed fixes out: bug reports, pull requests and issues, with Codex doing the heavy lifting and you merging.
Seven kinds of work your dot can take off your plate. Each opens the library filtered to that category.
Keep your day, your inbox and your loose ends in order. Includes morning brief from inbox and calendar and catch forgotten invoices.
Hand your dot a responsibility and let it track the moving parts. Includes keep an offsite on track and compare venues on a budget.
Watch sources, notice changes, and bring back only what matters. Includes competitor pricing watch and research a topic and report.
Proposals, accounts and follow-ups that stay current. Includes account follow-up drafts and keep a sales proposal current.
Turn raw material into drafts you approve. Includes content kit from an interview and keep launch materials aligned.
Feedback in, reviewed fixes out, with Codex doing the heavy lifting. Includes fix reported bugs for review and open pull request digest.
Errands, plans and reminders for life outside work. Includes bills and renewals reminder and plan a trip within budget.
What early users are reporting on X, credited and quoted. Their words, not confirmed by OpenAI.
i got to try openai dots a week ahead of the launch as part of their early access community … it's closer to Muse than Grokbot. one bot holding all your context, not multi-agent coordination … the interface feels like iMessage: chat or call your dots. SMS only for now, RCS and iMessage coming soon … now you invite your dot into Slack or Teams and it's already there
A pre-launch user's read: dots are one agent that holds your context, not a team of bots, and meeting you inside Slack or Teams is the real change. The note on SMS now and RCS and iMessage later is this user's report; OpenAI's docs describe texting only as a limited US beta.
The VM specs are modest: Debian Linux, 10 GB of memory, a 32 GB disk. It comes with a lot of serious productivity software preinstalled, such as Blender and FreeCAD. You can operate the VM by hand, or let your dot connect to your own computer.
A user's look inside the dot's cloud computer. The specs and preinstalled software are this user's report; OpenAI hasn't published them. OpenAI does document that you can take over the cloud computer yourself and optionally connect one personal computer.
New to dots?
Dots are OpenAI's always-on agents, each with its own cloud computer. They're rolling out gradually, and not every plan or region has them yet. Our reference guides cover setup, custom rules and what it costs, every fact linked to OpenAI.
Dots are always-on agents in ChatGPT, launched by OpenAI on September 29, 2026. Your dot has its own cloud computer, works across the apps you connect, remembers context, and keeps making progress between conversations, checking in with you on decisions that need your judgment.
Dots are rolling out gradually to Pro users outside the European Economic Area, Switzerland and the UK, and to Business Premium users in all supported ChatGPT regions. Enterprise workspaces can try a beta once an admin turns it on. You must be 18 or older.
Your first dot is included in Pro and Business Premium at no extra cost. Your plan also includes an allowance for deeper work, with extended limits for the first month after launch (the release notes word it as dots usage not counting toward plan allowances). Usage terms after that aren't published yet.
No. You create your dot in the ChatGPT desktop app (also available on Windows) or in ChatGPT on a desktop browser. After that, you can talk to it in the ChatGPT mobile app. Mobile web is not supported.
No. dots101 is an independent guide and is not affiliated with or endorsed by OpenAI. Every product fact links to OpenAI's own documentation, and every page shows the date we last checked it.