Terminal (AI inside)
For the complete documentation index, see llms.txtThe viewer has a docked terminal (toggle ⌃`) so you can drive the CLI — and your AI agent — without leaving the window.
Sessions
| Session | What it is |
|---|---|
| local | a native shell on your machine |
| claude | a Claude Code session, in the viewer |
| codex | a Codex session |
Open multiple tabs, switch between them, clear the buffer, and resize the dock. Each session is its own PTY over a dedicated WebSocket.
If the panel says it can’t start
The terminal is built on node-pty, a native addon that has no prebuilt binary
for every platform (linux-arm64 among them). It is an optional dependency:
on a machine without a C/C++ toolchain it is skipped rather than failing your
install, and the panel says so instead of hanging.
Everything else — runs, scenarios, the editor — is unaffected. To get the
terminal, install a toolchain (xcode-select --install on macOS,
apt install build-essential python3 on Debian/Ubuntu) and reinstall your
project’s dependencies.
Why it matters
Ask the agent to write or fix a test right here — then watch it run in the same window. The whole loop (author → run → inspect → repair) stays in one place.