This assistant works from one document, your board playbook, and does four things with it: classify, label, attach a checklist, ask for what is missing. It is a focused loop against a written spec, which is the kind of thing OpenCode is easy to aim at. Being model agnostic helps too, because playbooks grow from half a page to two without anyone deciding to let them.
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.
New site build: Kaselo Bakery, six pages, launch 14 Oct
added by Priya Raghunathan
Client is Kaselo Bakery in Lisbon, two shops. Scope agreed on the call: six page marketing site plus an online order form. Budget signed off at 8,400 EUR. They want it live by 14 October because the second shop opens that week. Day to day contact is Nuno Cardoso, nuno@kaselo.pt. Brand assets and photography are in the shared folder linked on this card.P
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect Trello
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.
Board playbook
Your board in your own words - which list is the intake, what kinds of request land there, the labels and checklist each type gets, where a card moves next, what counts as a complete request, and which cards are worth 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.
No exploration and no side quests. The playbook names the types, the labels and the checklists, and the agent executes against exactly that.
A longer playbook with more exceptions asks more of the model. OpenCode lets you move to a stronger one without rewriting a line of your rules.
Yes. Duplicate the workflow, set the copy to a different harness, and point it at a second intake list or a copied board. Because the playbook is a variable rather than code, both copies read the same rules and you are genuinely comparing harnesses instead of two sets of instructions. Keep only one of them pointed at the live intake list.
Free to start. Guided setup, a test run against staged conversations, and it's live.