Threads Mentions Agent with OpenCode
Voice is the product here, and voice is a model property far more than a harness one. OpenCode is model-agnostic by design, so the register of your drafts becomes something you can change directly: read a week of them, and if they sound like a help desk rather than like you, move the model instead of rewriting the voice notes.
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.
does this work if half the team is on Android, the last one we tried did not
The fourth reply in a thread somebody else started about tools for small design studios, now 61 replies long. She has 4,200 followers, runs a three person studio in Bilbao, and has never mentioned us before. The reply directly above hers says our onboarding was the easiest of the four she tested.
Set up in minutes
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect Threads
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.
Voice notes
How the account sounds and what it bothers replying to, in your own words: paste two or three real replies you were happy with, name the subjects you stay out of, and say who signs off on anything sensitive.
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
Register follows the model
Your pasted examples set the target and the model decides how close it lands. When drafts feel stiff, the model field is the honest first thing to change.
Voice notes stay put
Swapping the model leaves your examples, your no-go subjects and your sign-off rule untouched, so the comparison is genuinely like for like.
Before you fork
How do we tell a good draft from a plausible one?
By whether you would send it without editing, which is the standard the template writes to. Mid Thread Question is the useful test, since the right draft answers her Android question, acknowledges the reply sitting above hers, and stays inside the character limit. A draft that reads beautifully and answers something nobody asked is the failure that looks like success.
Run it with a different agent
Put Threads Mentions Agent to work on OpenCode
Free to start. Guided setup, a test run against staged conversations, and it's live.