Is Incident Update Drafter free?

The setup guide and prompt are free. In Claude it works on any plan, including Free. ChatGPT Work on the web needs Plus, Pro, Business, Enterprise or Edu, and Gemini tasks need Google AI Pro or Ultra.

Does it publish the update or change our status page?

No. It returns a draft, a fact-check map and approval questions. Your incident and communication owners approve every claim, and you publish through your usual status page or email tool.

Can it write a customer email instead of a status page note?

Yes. Set Update format to Customer email and it returns a subject and a short message with the same approved facts and next step.

Does it watch the incident or confirm recovery?

No. It works only from the facts you paste, connects to no monitoring and runs no schedule. A fix being deployed is not reported as recovery until you supply that fact.

Incident Update Drafter

Draft a customer incident update from approved facts

Incident Update Drafter is an AI agent you set up in the ChatGPT, Gemini or Claude you already use. Paste the approved facts about a service incident and it drafts a short status page note or customer email that keeps the affected scope, the unknowns and the next update time clear, with a fact-check map for your owners.

Pick your AI for the steps and the prompt

About each AI:ChatGPT WorkGemini TaskClaude app

Checked 29 Oct 2026. Not run by JuicyAgents yet. Written from each AI’s official docs; the example is illustrative.

Incident Update Drafter · reply in the chat

Status page note for INC-7, ready for approval

TitleCSV exports failing for some Region A customers
UpdateSome customers in Region A can’t export CSV reports. Viewing dashboards still works. Our engineers are investigating, and we don’t know the cause or when exports will work again yet. Our next update is at 11:00 UTC.
Fact checkEach line tied to F1, F2 or F3; F4’s vendor guess left out
To approveAny workaround or status link to add

Illustrative example: fictional incident INC-7 at 10:20 UTC. Some Region A customers can’t export CSV reports, the cause is unknown, and the next update is due at 11:00 UTC.

How it works

What it does for you, and what it leaves to you.

Service operators describe difficult choices about what to tell customers after an outage, while recipients report ignoring notifications that contain no useful information. Broad provider alerts may not reflect actual customer impact.

An assistant may help turn technical facts into readable updates. It cannot determine who is affected, confirm recovery or restore customer trust by itself. 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.

You stay in charge. 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.

You get

A customer update draft with a fact-check map and unresolved publication questions

It never

Publish the update, send it or change your status page
Promise a recovery time, or a next update time you haven’t approved
Add a cause, workaround, link or compensation that isn’t approved
Decide breach notices, liability or service credits

The update it drafts

The example above, in words.

Title: CSV exports failing for some Region A customers. Update: Some customers in Region A can’t export CSV reports. Viewing dashboards still works. Our engineers are investigating, and we don’t know the cause or when exports will work again yet. Our next update is at 11:00 UTC. The 11:00 time is a next update, not a recovery promise. The fact-check map ties each line to F1, F2 or F3 and leaves out F4’s vendor guess. The approval questions ask whether a workaround or status link can be added.

Settings

The setup prompt starts with the defaults, marked here.

Update format

Choose how the assistant organizes the result; evidence and review rules stay the same.

Status page note (default)
Customer email

Your working rules

Your approved tone, public service names and communication rules. Exclude internal secrets, personal data and unapproved technical details.

Optional. Empty unless you fill it in.

Check the first result

Each setup page has a fictional exercise to paste first. A good answer to it meets all of these.

11:00 UTC is a next-update time, not a recovery promise.
Removing an essential input causes a question or a blocked item, not an invented fact.

Questions people ask

Is Incident Update Drafter free?: The setup guide and prompt are free. In Claude it works on any plan, including Free. ChatGPT Work on the web needs Plus, Pro, Business, Enterprise or Edu, and Gemini tasks need Google AI Pro or Ultra.
Does it publish the update or change our status page?: No. It returns a draft, a fact-check map and approval questions. Your incident and communication owners approve every claim, and you publish through your usual status page or email tool.
Can it write a customer email instead of a status page note?: Yes. Set Update format to Customer email and it returns a subject and a short message with the same approved facts and next step.
Does it watch the incident or confirm recovery?: No. It works only from the facts you paste, connects to no monitoring and runs no schedule. A fix being deployed is not reported as recovery until you supply that fact.

Have us set it up

Rather not set it up yourself?

We set up Incident Update Drafter in your own accounts and the AI you already pay for, test it on your real work, and hand it over with a short walkthrough.

From $299

Runs in your accounts; you own everything
Tested before handover
Money back if we can’t make it work

Request a setup