OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

Tuui

Tuui is a desktop chat client that connects Model Context Protocol servers with configurable LLM backends, allowing users to work with MCP tools, prompts and resources while orchestrating model APIs from different providers.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tuui — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Tuui when you want a desktop workspace for connecting configured LLM backends to MCP servers, or to organize multiple model endpoint configurations.

Who it is for

  • Desktop users connecting MCP servers to LLMs with tool or function calling.
  • Developers configuring multiple chatbot endpoints or working with MCP server runtimes.

Documented uses

  • Connect an LLM backend to MCP servers and use their tools, prompts and resources.
  • Configure one or multiple chatbot endpoints from different providers.
  • Explore MCP server discovery and MCP Bundle support in a desktop client.
Before you choose

Requirements & limits

The README describes the application as requiring zero accounts and identifies release builds for Windows, Linux and macOS. MCP features still require a compatible LLM backend and any runtime dependencies needed by configured servers.

  • The selected LLM backend must support tool or function calling for MCP-related features.
  • MCP server prerequisites vary by runtime: Node.js for NPX/Node, Python and UV for UV/UVX, or Docker for Docker-based servers.
  • The project notes that macOS and Linux may require MCP configuration changes such as CLI paths or permissions.
  • LLM endpoint and MCP server configurations are separate; imported or modified configurations are stored in localStorage by default.
  • The documented MCP feature list includes tools, prompts, resources, sampling, elicitation, discovery and MCP Bundles; Roots is listed as unimplemented.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare a local TUUI development environment

Clone TUUI, install its dependencies, and start its Vite development environment.

Before you start

  • Git and Node.js/npm, or Yarn or pnpm, installed.
  • A terminal and a compatible desktop development environment.

Steps

  1. Clone the repository with the command below.

    Documented command
    git clone https://github.com/AI-QL/tuui

  2. From the cloned project directory, install dependencies with the command below, yarn install, or pnpm i.

    Documented command
    npm i

  3. Start the development environment with the command below.

    Documented command
    npm run dev

  4. If the application does not appear, check whether another application is using Vite’s default port, 5173.

Expected outcome

The project is set up and its Vite development environment is started.

Guide limitations

  • This recipe covers development setup and startup, not using MCP tools or configuring an LLM backend.
  • If port 5173 is occupied, the guide advises checking the port but does not document an alternative port procedure.
Documentation 1 ↗
Source-based practical guide

Set up an LLM endpoint configuration for TUUI

Prepare a single-chatbot or multi-chatbot JSON configuration for an LLM backend that supports tool or function calling.

Before you start

  • An LLM backend that supports tool/function calling.
  • The backend’s API key, endpoint URL, path, and model identifiers, as applicable.

Steps

  1. Choose whether the configuration will be a JSON object for one chatbot or a JSON array for multiple chatbots.

  2. Create an entry using the documented fields: name, apiKey, url, path, model, modelList, maxTokensValue, and mcp.

  3. Use the documented Qwen example as a single-chatbot template, or the OpenRouter and DeepInfra entries as examples for a JSON array.

  4. Place or import the configuration through the project’s documented LLM configuration path, src/main/assets/config/llm.json; imported or modified configurations are stored in localStorage by default.

Expected outcome

A JSON configuration is prepared for one or more LLM chatbots, with MCP enabled in the documented examples.

Guide limitations

  • The source gives configuration examples but does not document the in-app procedure for importing a configuration.
  • MCP server execution has additional runtime requirements: Node.js for NPX/NODE servers, Python and UV for UV/UVX servers, and Docker for Docker-based servers.
  • On macOS and Linux, the default MCP configuration may need CLI path or permission changes.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
AI-QL/tuui · Official homepage title · retained in its source language
How we review sourcesSuggest a correction