TUUI - Tool Unitary User Interface

TUUI - Tool Unitary User Interface is an open source desktop MCP client designed for tool integration and cross-vendor LLM API orchestration. It functions as a local AI playground that connects directly to various Model Context Protocol servers and compatible language model backends, such as ChatGPT, Claude, Qwen, OpenRouter, and DeepInfra. Developers, researchers, and AI practitioners use TUUI to execute function calls across modular tools, inspect prompts, access context resources, and test remote or local MCP servers without needing centralized account registrations. The application features built-in support for MCP tools, prompts, resources, sampling, elicitation, server discovery, and MCP bundles (.mcpb). Built with Vue 3 and TypeScript, it runs locally on Windows, macOS, and Linux, giving users control over API endpoints and configuration files while facilitating multi-model interactions within a unified chat interface.

Category: AI & LLM Tooling

Tags: client, interface, llm, orchestration

Visit TUUI - Tool Unitary User Interface

How to install and configure TUUI - Tool Unitary User Interface

1. Download the latest binary for your operating system (Windows, Linux, or macOS) from the project releases page at https://github.com/AI-QL/tuui. 2. Ensure prerequisites are installed on your host system: Node.js for NPX/Node-based tools, Python and uv for UV/UVX-based tools, or Docker for containerized servers. 3. Configure your LLM backend endpoints by editing resources/assets/config/llm.json or through the desktop interface settings, providing your API keys and compatible model parameters. 4. Configure local and remote MCP servers by editing mcp.json. For remote MCP servers using Cloudflare, add the server definition: json { "mcpServers": { "cloudflare": { "command": "npx", "args": ["-y", "mcp-remote", "https://YOURDOMAIN.com/sse"] } } } 5. Launch the desktop application and manage active tools and prompts directly within the interface.

What you can do with TUUI - Tool Unitary User Interface

• Orchestrating cross-vendor LLM APIs dynamically across OpenAI, Anthropic, Qwen, and DeepInfra backends in a unified interface. - Connecting to remote and local MCP servers via NPX, UV, and Docker environments for tool and function calling. - Discovering and testing community MCP tools, prompts, and resources directly through Model Context Protocol registry integration. - Running MCP Bundles (.mcpb) and executing automated application testing routines within a local desktop environment.

Key facts

• https://github.com/AI-QL/tuui
• AI & LLM Tooling, Developer Tools & Code Intelligence
• client, interface, llm, orchestration

Part of MCP Servers

What is TUUI - Tool Unitary User Interface?

TUUI is an open source desktop Model Context Protocol client and AI playground. It allows users to run local and remote MCP tools, organize prompt workflows, access dynamic context resources, and interact with multiple third-party LLM providers from a single interface without third-party account requirements.

Which operating systems support TUUI?

TUUI provides release packages for Windows, Linux, and macOS platforms. Host environments may require additional runtime utilities such as Node.js, Python with UV, or Docker to properly launch and spawn specific local MCP servers.

What MCP features does TUUI support?

TUUI supports MCP Tools, Prompts, Resources, Client Sampling, Elicitation, dynamic server Discovery on the MCP registry, and MCP Bundles (.mcpb extensions). It coordinates these capabilities directly with configured tool-calling language models.

How do I configure remote MCP servers in TUUI?

Remote MCP servers can be configured in the mcp.json settings file by executing the npx mcp-remote command with the remote SSE endpoint URL. TUUI can then handle authentication redirects and proxy server capabilities directly into the chat session.

Is TUUI - Tool Unitary User Interface open source?

Yes, TUUI is open source software hosted on GitHub, built with Vue 3, Vuetify, and TypeScript. Developers can download pre-built releases or inspect and modify the source code under its project repository.