Terminal
Open a live shell inside your Agent's sandbox and run commands directly from the browser.
The Terminal tab gives you a real shell inside the Agent's cloud computer. You can run commands, install packages, inspect files, and debug scripts — all without leaving Buda.
Opening a terminal
Click the + button in the tab bar and select Terminal. The tab connects to the sandbox within a few seconds. If the sandbox is asleep, Buda wakes it automatically — the first connection may take 10–20 seconds while the environment starts.
Running commands
Type any shell command and press Enter. The terminal supports a full POSIX shell:
# Navigate the file system
ls -la
# Run a script
python3 scripts/process_data.py
# Install a package
pip install pandas
# Check running processes
ps auxThe working directory starts at the Agent's Drive root, so files you create here show up immediately in the Files tab and vice versa.
Multiple panes
Open more than one terminal to run things side by side — click + and select Terminal again. Each pane is an independent shell session. Close a pane with the × in its corner.
Persistent sessions
The session lasts as long as the sandbox is running. Switch tabs and come back — your shell history and running processes are still there. When the sandbox sleeps after its idle timeout, the terminal reconnects automatically on your return (see Runtime monitoring).
Environment variables
Variables you set in the Agent's settings are available in the terminal automatically (see Environment variables). You can also export them inline for the current shell:
export API_KEY=your_key_here
python3 run.pyWatching the Agent use the terminal
When the Agent runs commands as part of a task, you can see the exact commands and output inline in the Agent tab. Switch to the Terminal tab to inspect the live shell state afterward.
Tips
Ctrl+C— interrupt a running commandCtrl+L— clear the screenTab— auto-completion- ANSI colors and most interactive CLI tools are supported