Drafting a support reply is mostly reading: the whole thread, the Notion page that already answers it, the policy you wrote six months ago. Claude Code is deliberate about gathering context before it does anything, which is exactly what this agent needs, because a draft written from a page title instead of the page itself is how wrong refund windows get promised. It carries the fetch, the searches and the writing as one connected task rather than answering from memory.
Run a staged conversation — no account needed. The agent handles it for real while a simulated world answers its tool calls; nothing touches real accounts, and nothing is actually sent.
Adding two more seats before our audit on the 30th
Good morning, we are a six person accounting studio in Milan and we are on your team plan. Two colleagues join us on Monday and I need them in the account before a client audit on 30 August. Can I add the seats myself from the billing page, and does the price change straight away or at the next renewal? One more thing, will they be able to see the archived files from last year?
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect Gmail
One sign-in. The agent acts through your account, scoped to what this template uses.
Connect Notion
One sign-in. The agent acts through your account, scoped to what this template uses.
Business name
The name your customers know you by, used in the greeting and sign-off of every drafted reply.
Support playbook
Your support policies in your own words - refund and cancellation rules, what you can and cannot promise, and the tone and sign-off you want on every reply. The agent may only answer from this and your knowledge base.
Kb notes
What lives in your Notion knowledge base and how to use it - which pages hold product docs, policies, and troubleshooting, plus anything in there that is out of date or off limits. The agent searches Notion before it writes a word.
Runs on Claude Code
Preselected for this page — connect your Claude Code account during setup, or switch to NoClick's built-in models with one click.
Watch it handle a test run
A staged conversation against a simulated world — then it’s live.
The agent is told to pull the full thread and open the Notion pages rather than trust their titles. Thorough context gathering is the part of Claude Code that carries this workload.
One email can mean a thread fetch, several knowledge base searches, and a draft that has to end with your sign-off. Long structured sequences are where this harness stays reliable to the last step.
Yes. Claude Code runs on your credentials, so during setup you either sign in with Claude or paste an Anthropic API key. Model usage bills to that account, and NoClick charges for the hosted sandbox time the agent spends on each ticket.
Free to start. Guided setup, a test run against staged conversations, and it's live.