What exactly does the OpenCode integration do?

It installs a hook in your OpenCode config. When OpenCode stops, finishing a turn or asking a question, wispa shows a reply window where you answer by voice or text, and sends your answer back into the session.

Do I have to edit config files by hand?

No. Flip the OpenCode switch in wispa's Integrations tab and it writes the hook for you; toggle it off to remove it cleanly. Start a new OpenCode session for the change to take effect.

Does it pop up on every single message?

By default wispa only opens when OpenCode's message reads like a question or a request for input, and stays quiet during ordinary build output. Prefer it on every turn? That's a single toggle.

Does my answer go back into the same session?

Yes. Continuation mode returns your reply straight into the running session and works even when the terminal is in the background. You can also choose Paste, which types it into the terminal like a normal dictation.

Can I fix a mistake before it sends?

Always. The window opens with a text field, so you can speak or type, correct anything, and nothing is sent until you press Send or Enter. Esc dismisses it without answering.

Does this work with Claude Code, Codex, Gemini CLI, Cursor and Pi?

Yes. The same reply overlay supports Claude Code, Codex, Gemini CLI, Cursor and Pi too, each with its own page. This one covers OpenCode.

wispa for coding agents

Answer OpenCode by voice.

OpenCode is the open source agent you run in your own terminal, against your own models. It works in long build loops, then stops when a decision is yours to make. wispa catches that moment: a reply window opens wherever you are, you answer by voice, and it goes straight back into the session.

Download for Mac Download for Windows

Requires wispa for macOS 13+ or Windows 10+. Free in beta, no account.

OpenCode is waiting

wispa-landing/auth

The useSession hook is duplicated across three components, each fetching a little differently.

Want me to:

Extract one shared provider and consume it everywhere
Keep them separate but dedupe the fetch logic
Leave it and just fix the stale-token bug

Which direction should I take?

Cancel esc Speak ⌘L Send ⏎

How it works

Set it up once, then just talk.

Turn on the OpenCode hook

Open wispa, go to Integrations, and flip the OpenCode switch. It writes the hook for you; start a new session to activate.
OpenCode pauses and asks

When your agent needs you, the reply window appears over whatever app you are in, with its message and project.
Answer out loud

Hold your dictation key and speak, or just type. Hit Send, and your answer lands back in the session.

Why it's better

Built for the way agents actually interrupt you.

Leave the TUI in the background

OpenCode runs in a full terminal UI, but you don't have to sit in it. The moment build mode stops for a decision, wispa surfaces over your editor with the full message, so you never miss the prompt behind another window.

Long directions, spoken

OpenCode takes rich, paragraph-length direction, and speaking it is far faster than typing. wispa sets your words down as clean text right in the reply box, then sends it on.

Back into the running session

Continuation drops your answer straight into the live OpenCode session, even when the terminal sits in the background, no matter which model OpenCode is driving. No window juggling, no lost context.

FAQ