Use case · Support
Turn the questions customers keep asking into help articles
To write help center articles from support tickets with AI agents, connect your help desk, such as Zendesk, Intercom, Help Scout, or Freshdesk, to a Brainwrite bot. It reads recently solved tickets, groups repeated questions, and checks your existing articles for gaps. A writer bot drafts an article for each gap from the answers that resolved the tickets. You review and publish.



Every Friday, read this week's solved Zendesk tickets. Group them by question, tell me which questions came up three or more times with no help article, and draft an article for each one in the Help drafts folder in Google Docs. Remove customer names and order numbers from examples.
Your Chief of Staff put a team on it
Nova· Chief of Staff
Atlas· Ticket analyst
Sage· Article writer
- 214 solved tickets read; 9 questions came up 3 or more times
- Missing article: 'How do I change the billing email?' (17 tickets this week)
- Outdated: the export article describes the old menu; 6 tickets mention the new one
- 3 article drafts saved to Help drafts, customer details removed
DoneSaved as a routine: Fridays at 14:00. It starts paused until you turn it on.

Help center articles from tickets: done. Nothing sent.
- 1214 solved tickets read; 9 questions came up 3 or more times
- 2Missing article: 'How do I change the billing email?' (17 tickets this week)
- 3Outdated: the export article describes the old menu; 6 tickets mention the new one
The job
Doing it yourself, and handing it off.
Doing it yourself
The same question comes in five times a week, and each time an agent writes the answer again, slightly differently. Everyone agrees there should be an article. Nobody has time to write one between tickets, and nobody knows which questions are the most common, because the help desk counts tickets, not topics. The help center slowly drifts out of date while the best answers sit in solved tickets.
What the team does
- Read solved tickets and group them by the question customers asked
- Count how often each question came up in the period
- Check existing help articles, solution articles, or saved replies for coverage
- Draft new articles from the answers that resolved the tickets
What waits for you
- Which articles get published, and where
- Whether an answer reflects current policy and product behaviour
- Any change to tickets, macros, or saved replies
The team
Who's on it.
Your Chief of Staff picks the specialists the job needs. Each has its own role, instructions, model, and app access.
NovaIllustrated exampleChief of Staff
Runs the weekly review and hands you the list of gaps and drafts.
AtlasIllustrated exampleTicket analyst
Reads solved tickets in the help desk, groups repeated questions, and checks existing articles or saved replies for each one.
SageIllustrated exampleArticle writer
Drafts a help article for each gap from the replies that resolved the tickets, in your saved style. Drafts only.
How it runs
A run, step by step.
- 01
Connect the help desk
Connect Zendesk, Intercom, Help Scout, or Freshdesk, and grant the analyst read tools only.
- 02
Find the repeats
The analyst groups solved tickets from the period you choose by question and counts each group.
- 03
Check for gaps
It compares each group with your existing articles and marks it covered, outdated, or missing.
- 04
Review the drafts
The writer drafts articles into Google Docs, Notion, or Confluence. You edit and publish in your help center.
What you'll connect
Apps this job uses.

Zendesk
Triage Zendesk tickets, read full threads, draft replies with your macros, and merge duplicates you confirm.

Intercom
Search Intercom conversations, draft replies, add notes, tag and assign chats, and keep help articles current.

Help Scout
Read Help Scout conversations, draft replies from saved replies, tag and note threads, and summarise inbox reports.

Freshdesk
Filter Freshdesk tickets, add private notes, draft replies from canned responses, and summarise satisfaction ratings.

Confluence
Bots search Confluence, read pages and comments, and create or update pages and blog posts.
FAQ
Questions about Help center articles from tickets
Can the bot publish articles straight to my help center?
The Intercom connector can create and update help center articles, so a bot with those tools could write directly into it. Keep those tools off, or keep the thread in Ask mode, until you have reviewed the text. Drafting into Google Docs, Notion, or Confluence first keeps publishing a clear human step.
Will customer details end up in the articles?
Tell the writer to remove names, emails, and order numbers from every example, and save that rule in its memory. Ticket text the bots read is still sent to the AI provider you chose, so pick one you trust with support data, and review each draft before it is published.
Which help desks does it work with?
Zendesk, Intercom, Freshdesk, Help Scout, Gorgias, and Zoho Desk connectors can all read ticket conversations. Freshdesk can also read solution articles, Help Scout saved replies, and Intercom help center articles, which helps the bot check what is already covered.
Built on
Features this job relies on.
Connected apps
Gmail, GitHub, Slack, Notion and hundreds more via OAuth, with per-bot tool grants.
Routines
Run a bot's task once, on an interval, or on chosen weekdays, with a receipt for every run.
Memory
Plain-text memory per bot that you can read, fix and delete. No hidden profile.
Approvals
Risky actions become cards you approve or deny in the conversation. Ask or Full Access per thread.
Give your first job to Brainwrite.
Download the app, connect the AI you already pay for, and tell your Chief of Staff what needs doing.
macOS today. Windows and Linux are coming soon.







