Zendesk Ticket Triage Agent with OpenCode
Triage is the highest-volume job in a support stack, one run per ticket indefinitely, so the cost per ticket decides whether you can afford to leave it on. OpenCode lets you choose the model behind it and change your mind later without touching the rules, the tags or the Slack channel. Classification quality is the thing to watch while you tune that.
Watch it work before it's live
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.
Checkout returns a 500 error for every customer since 08:15
OpenAnneke van der Meer <anneke@bloemhaven.nl>
Since 08:15 this morning every customer who clicks Pay gets a 500 error page and no order is created. I have tested in three browsers and on two devices, and it reproduces on our staging store as well. We are a florist and this is our peak delivery window, so we are losing roughly 40 orders an hour. Nothing changed on our side except the platform update notice you sent yesterday.
Set up in minutes
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect Zendesk
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.
Triage rules
How you sort tickets - your categories, the tags each one gets, which queue or person owns it, and what to do with a ticket that fits nothing. The agent classifies against this and nothing else.
Urgent criteria
What makes a ticket worth interrupting someone for - the symptoms, customers, or words that mean now rather than tomorrow. Only tickets matching this reach Slack.
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.
Why OpenCode for this agent
Volume makes the model matter
At a thousand tickets a week the model choice stops being a detail. Being able to set it is the difference between triaging everything and triaging a sample.
Rules outlive the model
Your categories, tags, owners and urgent criteria live in the workflow rather than the harness, so a model swap later costs you a test run and nothing else.
Before you fork
How do I tell if the model is classifying badly?
The internal note shows its working, the suspected cause and the questions it would ask, so a wrong call is visible instead of silent. Spend the first week reading notes on tickets whose answer you already know. If the cause line is consistently vague, that is a model signal rather than a problem with your rules.
Run it with a different agent
Put Zendesk Ticket Triage Agent to work on OpenCode
Free to start. Guided setup, a test run against staged conversations, and it's live.