Sorting mail into categories you already defined is not the hardest thing a language model does, which is precisely why picking the model is worth something here. OpenCode lets you run triage on something modest, then move up only if labels start coming back wrong. Your routing rules, labels and owners are written once and survive whatever sits underneath.
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.
All 34 stores offline since 09:40
Our tills stopped syncing at 09:40 this morning and all 34 stores are taking cash only. We restarted the local gateway in two stores with no change, so it looks like the problem is on your side. Saturday is our busiest day and we are losing roughly 40,000 kroner an hour. Please call my mobile, I am reachable all day.
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 Slack
One sign-in. The agent acts through your account, scoped to what this template uses.
Routing rules
Your categories, the exact Gmail label for each one, who owns it, and the fallback label and owner for mail that fits nothing. The agent never invents a label or sends to anyone outside this list.
Urgent criteria
What genuinely cannot wait, such as the senders, deadlines, amounts, or phrases that should interrupt someone, plus the channel that has to hear about it. Everything else is labeled and forwarded without a Slack ping.
Runs on OpenCode
Preselected for this page — connect your OpenCode 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.
A bounded choice among labels you wrote does not demand a frontier model. Model freedom turns that into money saved on every single email.
The staged emails give you the same three cases against any model, so downgrading is something you verify rather than hope about.
Yes. The model is a field on the agent, while your routing rules, urgent criteria, Gmail connection and Slack channel live separately from it. Re-run the staged emails after any change, since a smaller model tends to show its limits on the ambiguous one first.
Free to start. Guided setup, a test run against staged conversations, and it's live.