Is Fix Confirmation 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.

Does it connect to Zendesk or Jira?

No. You paste the tickets and the fix note as text. Helpdesks such as Zendesk can already update a ticket when its linked Jira issue changes status. This covers what that skips, whether the fix reached each customer and what to tell each one.

Does it send the replies or close tickets?

No. It returns a table with one row per ticket, a draft for each customer and notes for engineering. Support checks each draft, sends it and changes ticket states by hand.

What if a customer still has the problem after the release?

Paste their reply into the same chat and run it again. A customer who reports the problem after the release is marked reopened, whatever the deployment says, with a note for engineering.

How many tickets can I paste at once?

Up to twenty tickets for one bug. Use a new chat for each bug, so customers stay apart.

Fix Confirmation Drafter

Tell each customer when their bug is really fixed

Fix Confirmation Drafter is an AI agent you set up in the ChatGPT or Claude you already use. Paste the tickets for one bug with the engineering fix note, and it checks customer by customer whether the fix reached them, drafts a reply for each, and keeps open every case that can’t be told it’s fixed.

Pick your AI for the steps and the prompt

About each AI:ChatGPT WorkClaude app

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

Fix Confirmation Drafter · reply in the chat

BUG-81 follow-ups, 1 ready for a customer check and 3 kept open

ZD-601Ready for a customer check: a draft asks Cedar to retry the 7,000-row export and confirm
ZD-602Waiting for the customer: the draft asks Pine for the row count, error code and steps
ZD-604Waiting for engineering: Oak is on US production with export_v2 off
ZD-605Reopened: Birch still sees E42 after the release

Illustrative example: four fictional tickets about CSV export error E42, an engineering note saying the fix shipped to EU production behind the export_v2 flag, and one customer who still sees the error.

How it works

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

A team announces a fix while some customers still have the problem: the release reached one environment, a feature flag is off for their account, or their report was a different bug. The promised update slips because nobody tracks each customer separately once the engineering issue is marked done.

Ticket tools already link a bug to its engineering issue and update the ticket when the issue closes. What they do not do is decide, customer by customer, whether this person’s environment, version and feature flag actually got the fix, and write a reply that fits their history. The assistant works from text you paste, checks the release gates for each case, drafts a reply only where the fix applies, and keeps the rest open with a named next action. Nothing is sent or closed by the assistant.

You stay in charge. Engineering confirms the issue match and the release mapping; support checks every draft against the customer’s own ticket. Support sends replies and changes ticket states manually, and records each outcome.

You get

One row per customer: proposed state, the release gates that pass or fail, the next action and its owner
A separate draft reply for each customer who can be told, and a holding or follow-up draft for each who cannot, plus internal notes for engineering

It never

Send replies, close tickets or contact engineering
Tell a customer it’s fixed while a release check fails or is unknown
Write one message for everyone or mention another customer
Treat a customer’s silence as a fix

The follow-ups it prepares

The example above, in words.

ZD-601 is ready for a customer check: a draft asks Cedar to retry the 7,000-row export and confirm, with the 16:00 promised update noted as due in an hour. ZD-602 stays waiting for the customer: the draft asks Pine for the row count, error code and steps and makes no fix claim, with the 16:30 promised update also flagged as due within two hours. ZD-604 stays waiting for engineering: Oak is on US production with export_v2 off, so the draft is a holding note and the internal line asks Leon when US ships. ZD-605 is reopened: Birch still sees E42 after the release, so the draft acknowledges it and asks for the exact steps, and the internal note tells Leon the fix did not hold for this customer. No bulk message, no ticket closed, and the instruction inside the engineering note is ignored and reported.

Settings

The setup prompt starts with the defaults, marked here.

Reply style

Changes the wording of the customer drafts only.

Clear and friendly (default)
Brief and direct

Closure policy

What the assistant may propose when a customer does not reply. You still record every closure yourself.

Keep open until the customer confirms (default)
Propose closing after two unanswered follow-ups

Team rules

Optional wording, contact hours or approved policy. No customer data is needed here.

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.

ZD-601 is ready for a customer check with a draft that asks Cedar to retry and confirm; the 16:00 promised update is flagged as due within two hours.
ZD-602 stays waiting for the customer: the draft asks for the row count, error code and steps and claims no fix; the 16:30 promised update is flagged as due within two hours.
ZD-604 stays waiting for engineering: the gates show US production not deployed and export_v2 off, and the draft is a holding note with no fix claim.
ZD-605 is reopened despite the successful EU deployment, with an internal note to Leon.
No bulk message is written, no ticket is closed, no other customer is mentioned in any draft, and the instruction inside the engineering note is reported as ignored.
Removing an essential input causes a question or a blocked item, not an invented fact.

Buy instead

Ready-made products that do a similar job, if you would rather pay for one.

CannyCanny collects feature requests in one place, uses its Autopilot AI to pull requests out of support chats, sales calls and app reviews and merge duplicates, and emails voters when a request's status changes.

Questions people ask

Is Fix Confirmation 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.
Does it connect to Zendesk or Jira?: No. You paste the tickets and the fix note as text. Helpdesks such as Zendesk can already update a ticket when its linked Jira issue changes status. This covers what that skips, whether the fix reached each customer and what to tell each one.
Does it send the replies or close tickets?: No. It returns a table with one row per ticket, a draft for each customer and notes for engineering. Support checks each draft, sends it and changes ticket states by hand.
What if a customer still has the problem after the release?: Paste their reply into the same chat and run it again. A customer who reports the problem after the release is marked reopened, whatever the deployment says, with a note for engineering.
How many tickets can I paste at once?: Up to twenty tickets for one bug. Use a new chat for each bug, so customers stay apart.

Have us set it up

Rather not set it up yourself?

We set up Fix Confirmation 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