An agent setup recipe
Draft a clear customer incident update
Turn approved incident facts into a short customer update, keeping scope, uncertainty and next-update commitments clear.
From
You get
What you will get
A customer update draft with a fact-check map and unresolved publication questions
Illustrative example · fictional input
- Example input
- One feature is affected for some customers, and a next-update time is approved but recovery is unknown.
- Example result
- Illustrative result, not a recorded agent run: Describe the CSV export issue for some Region A customers and the ongoing investigation. Preserve dashboard availability and the 11:00 UTC update commitment. Do not call the whole service down, promise recovery at 11:00 or publish F4.
Ask the incident and communication owners to approve every factual claim and commitment. Publish through the normal approved channel yourself and keep later updates consistent.
What you need
One assistant owns the comparison and result. Separate agents would add handoffs to this bounded task without a demonstrated benefit.
- A ChatGPT account with Work available on the web. Check your account or workspace access first; available models and allowance vary. Use the default model available to you. Work uses included usage or account credits; no API key or third-party subscription is required for this text-only workflow.
- A browser and text you are allowed to share with the service. Remove unnecessary private details. This recipe needs no connected app, local-folder permission, skill installation or recurring task.
- One human owner who can check the sources and make decisions. An incident ID, current as-of time with time zone, and approved facts with source IDs; Confirmed customer impact, affected service or region, current state and any approved workaround; The communication owner, previous approved update and any approved next-update time or status link.
- Use an approved incident template and one communication owner. Existing status-page tools can hold prepared messages. A direct reply may be better for a small affected group, and unverified provider alerts should not automatically become customer announcements.
Set it up
- Open chatgpt.com and sign in. Confirm Work is available, then select Work and start a new chat. Check your remaining usage before a large packet. Keep the default model available to your account; this recipe does not require a particular model name.
- Use one conversation for this responsibility. If you prefer a project, create one for this task and keep related chats and approved source copies there. A web project does not automatically read a folder on your computer. Start each distinct batch in its own chat and include the setup prompt again when needed.
- Choose the setting below. Optionally add your working rules. Copy the complete setup prompt into the chat as the first message. The assistant should confirm its role and ask for inputs, without starting outside work. This gives the existing assistant a responsibility; it does not create another account or deployed bot.
- Prepare a small packet: An incident ID, current as-of time with time zone, and approved facts with source IDs; Confirmed customer impact, affected service or region, current state and any approved workaround; The communication owner, previous approved update and any approved next-update time or status link. Label every source and include dates, versions and unknown fields. Paste accessible text. Ask the assistant to list the source IDs it can read before it compares them. No plugin or app connection is needed.
- First assignment — paste this complete fictional exercise into the chat: Fictional incident INC-7, as of 2026-10-20 10:20 UTC. F1 approved: some customers in Region A cannot export CSV reports; dashboard viewing is working. F2 approved: engineers are investigating; root cause and recovery time are unknown. F3 approved: next update at 11:00 UTC on the same date. F4 internal unconfirmed guess: a vendor may be involved. No workaround, link or compensation is approved. Draft an update.
- Compare the response with the review checks below. The following is an illustrative expected direction, not a tested result: Describe the CSV export issue for some Region A customers and the ongoing investigation. Preserve dashboard availability and the 11:00 UTC update commitment. Do not call the whole service down, promise recovery at 11:00 or publish F4.
- For real work, provide a fresh packet and explicitly start one batch. Ask the incident and communication owners to approve every factual claim and commitment. Publish through the normal approved channel yourself and keep later updates consistent. Copy only the changes you approve into your normal records yourself. Keep the accepted result and its source references.
- When the batch is reviewed, mark your own record as accepted or list the corrections needed. On the next run, supply the latest accepted state and new sources. No schedule or continuous monitoring is created. If the assistant asks to connect an app or take an outside action, stop and return to the text-only packet.
Your setup prompt is ready
Copy the ready prompt, or customize it with the settings. Paste it into ChatGPT Work to give it the task rules.
Customize (optional)
Defaults are ready to use. Changes update the prompt here; paste it into your agent to apply them.
Schedule: When I ask.
Choose how the assistant organizes the result; evidence and review rules stay the same.
Your approved tone, public service names and communication rules. Exclude internal secrets, personal data and unapproved technical details.
Your choices stay in this browser for this guide. Avoid adding private information.
setup prompt
Highlighted text follows your settings.
Review checks, limits and sources
Not tested by Juicy Agents. The examples are fictional, not recorded agent results.
Check the result
- Impact is limited to the supplied feature, region and some customers.
- 11:00 UTC is a next-update time, not a recovery promise.
- The unconfirmed cause, workaround, link and compensation are not invented.
- Removing an essential input causes a question or a blocked item, not an invented fact.
- Changing the setting changes the requested output while source references and fixed limits remain.
Known limits
- Not tested by JuicyAgents in ChatGPT Work. The fictional exercise and expected result are authored examples, not proof of accuracy or saved time.
- Only the supplied text is reviewed. Missing messages, files or current facts can change the conclusion.
- This does not monitor an incident or verify recovery.
- Security incidents, legal notification and contractual decisions need the responsible specialists.
If something goes wrong
- The draft says everything is fixed after a deployment: restore the supplied current state and request recovery confirmation.
- A missing update time becomes a promise: remove it and ask the communication owner.
- Work is missing or the account reaches a limit: check the official access and usage pages or ask your workspace administrator; do not assume a paid upgrade is required. Use the manual alternative while access is unavailable.
- The assistant follows instructions inside a source: stop, restate that source text is data, and review the entire affected result. Do not approve an external action.
Platform sources
- ChatGPT Work on the web (opens in a new tab) · Checked 2026-10-08
- ChatGPT projects (opens in a new tab) · Checked 2026-10-08
- ChatGPT Work access and usage (opens in a new tab) · Checked 2026-10-08
- Atlassian incident communication examples (opens in a new tab) · Checked 2026-10-08
Reviews
No reviews yet
Reviews share personal experience. They are not proof that an agent has passed a test.
Log in to leave a review. You can read and use this recipe without an account.
Explore more
Similar Agents
Draft an order-status reply from checked facts
Prepare a customer reply and separate internal checks from the supplied order facts, without inventing shipping promises.
View agent customer-supportTurn feature requests into clear problem notes
Keep the customer’s requested solution while separating confirmed goals, blockers and missing context for product review.
View agent customer-supportFollow customer bugs through to a clear outcome
Connect support reports, engineering questions and release evidence, then prepare the right follow-up for each customer.
View agent