terminal-use runs on your own machine — the client starts it, so there is no endpoint to ping. 487 installs a week from npm. Last commit 8 Oct 2026.
Lets AI agents drive a real terminal: type, press keys, click, read the screen, take screenshots.
What the code does
We read the source, 2 d ago · rules ccca72095570
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
const session = this.#spawn(tomb.config)
for (const [k, v] of Object.entries(process.env)) {
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
This code can reach further than it looks
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Connect this server
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add terminal-use -- npx -y terminal-use{
"mcpServers": {
"terminal-use": {
"args": [
"-y",
"terminal-use"
],
"command": "npx"
}
}
}[mcp_servers.terminal-use]
command = "npx"
args = ["-y", "terminal-use"]{
"mcpServers": {
"terminal-use": {
"args": [
"-y",
"terminal-use"
],
"command": "npx"
}
}
}{
"mcpServers": {
"terminal-use": {
"args": [
"-y",
"terminal-use"
],
"command": "npx"
}
}
}Alternatives to terminal-use
same job, measured the same wayThe terminal AI agents can drive: spawn panes, run commands, read screens, scrolling screenshots.
Drive the browser of ZennoPoster ProjectMaker: open pages, read and click elements, screenshots.
Let AI agents see and interact with terminal/CLI apps via PNG screenshots
Drive the browsers of running ZennoPoster tasks: open pages, read and click elements, screenshots.
Let AI agents inspect and drive running WPF apps: tree, bindings, screenshots, clicks, text input.
Lets a coding agent read the screenshots you beam to its server with Hostbeam.
AI agents see and drive the Android Emulator: screenshots, UI in dp, tap, dark mode, RTL, logcat.
Control a local headless Chrome browser: open pages, read titles, and take screenshots.
terminal-use — questions
Answers built from our own checks of this server.