open-webui/open-terminal

What it solves

AI assistants are capable of writing code but often lack a secure, dedicated environment to execute it. Open Terminal provides a self-hosted, remote shell that allows AI agents and automation tools to run commands, manage files, and execute code via a REST API.

How it works

It operates as a lightweight server that can be deployed in two ways: as a sandboxed Docker container (which comes pre-installed with tools like Python, Node.js, and data science libraries) or as a bare-metal installation via pip. It exposes a REST API for command execution and file management, and can be configured with custom packages via environment variables. It also integrates directly with Open WebUI to provide a file navigation sidebar and terminal access for AI assistants.

Who it’s for

Developers and AI agent creators who need a remote execution environment for their LLM-based tools, as well as users of Open WebUI who want to give their AI assistants direct access to a system shell.

Highlights

• Flexible Deployment: Supports both isolated Docker containers and direct bare-metal installation.
• Curation of Tools: The latest Docker image includes a comprehensive toolkit (Node.js, gcc, ffmpeg, LibreOffice, LaTeX, etc.) for AI agents.
• API-Driven: Everything is accessible via a simple REST API, making it easy to integrate into automation workflows.
• Open WebUI Integration: Provides a seamless connection for AI assistants to browse and edit files and run commands.
• Multi-User Support: Offers a lightweight multi-user mode with per-user Linux accounts and home directory isolation.