Meeting Follow-up Agent with OpenCode
Live demo
Watch it work before it's live
Discovery call: Brillante Caffe Group
Recording finished, 38 minutes, with Elena Rossi, Chief Operating Officer at Brillante Caffe Group in Milan, and two people from our side. They run 41 cafes and stock ordering is done by each manager over WhatsApp, which they estimate cost about 6,000 euro in waste last month. Elena asked for pricing across 41 sites and an example rollout plan, and our rep said both would be sent by Friday 21 August. Elena said she would send a point of sale export on Tuesday, and both sides settled on a second call for 27 August at 14:00 CEST. She also asked whether the app works offline in their two mountain locations, which nobody on the call could answer.
What you'll need
Set up in minutes
Connect Fathom
One sign-in. The agent acts through your account, scoped to what this template uses.
Connect Gmail
One sign-in. The agent acts through your account, scoped to what this template uses.
Connect HubSpot
One sign-in. The agent acts through your account, scoped to what this template uses.
Followup style
How your follow-up emails read - length, tone, greeting and sign-off, whether next steps are bullets or sentences, and anything you always include or never include.
Crm rules
What you want logged in HubSpot after a call - which meetings deserve a note, what that note has to contain, and how much detail the next person opening the record will need.
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 considered pairing
Why OpenCode for this agent
Cost scales with talking
A long call is a long input, so cost per run tracks meeting length rather than headcount, and the model is your lever on that.
Room for the transcript
Choosing the model also chooses how much of a call fits comfortably in one run, which matters more in this template than in any other.
Good to know
Before you fork
Is a cheap model enough to write a follow-up?
It is usually enough for the recap and often not enough for the commitments, since pulling out who promised what by when takes closer reading than summarising does. Run the staged discovery call, which contains four separate commitments with names and dates, and count how many survive. Whatever goes missing there will go missing on real calls too.
Other editions
Run it with a different agent
Make it yours
Put Meeting Follow-up Agent to work on OpenCode.
Free to start. Guided setup, a staged test run, and a workflow you can change at every step.