Docs / Help

Frequently Asked Questions

Quick answers to the questions we hear most often.

How do I install CLI Launcher?

Run the one-line installer — curl -fsSL https://cli-launcher.veyl.in/install.sh | bash on macOS and Linux, or powershell -ExecutionPolicy Bypass -c "irm https://cli-launcher.veyl.in/install.ps1 | iex" on Windows. You can also download a native .dmg, .msi, or .AppImage from the Downloads page, or build from source. See Installation.

Which terminal AI agents can CLI Launcher manage?

CLI Launcher manages 34+ agents across three runtimes: 19 Node-based (Claude Code, OpenCode, Gemini CLI, GitHub Copilot, Codex, Oh My Pi, and more), 7 Python-based (Aider, Open Interpreter, Shell-GPT, gptme, RA.Aid, OpenHands, Hermes Agent), and 8 standalone binaries (Goose, Cursor, Amazon Q, Plandex, Mods, aichat, Fabric, and others). The full list is on the Supported Agents page.

What is YOLO Mode and is it safe?

YOLO Mode launches supported agents with their auto-approve flags enabled (for example, Claude Code's --dangerously-skip-permissions) so they run unattended without permission prompts. Because it auto-approves every action, use it only in trusted working directories. More in YOLO Mode.

Does CLI Launcher sandbox the agents it installs?

No. CLI Launcher is a management panel, not a sandbox. It installs and launches agents using your system's existing Node, Python, or standalone runtimes; it does not create isolated environments or virtualize dependencies.

Is CLI Launcher free and open source?

Yes. CLI Launcher is MIT-licensed and free to use. The source code is available on GitHub.