convez.ai
MCP ServerAutomaçãoIntermediário
shulain

MobausStudio

shulain/MobausStudio

Cliente unificado para 20+ provedores de IA com MCP e orquestração de agentes

Ver no GitHub
InicialPluginScore · de 0 a 100

Cliente de IA multiplataforma open-source com suporte a 20+ provedores de IA, extensões MCP e orquestração de agentes. Integra ChatGPT Plus, Claude Pro, GitHub Copilot e outros via OAuth, sem necessidade de API keys. Oferece sistema de skills (30k+ templates), discussões multi-agente e interface unificada para todos os modelos.

Instalação

$npx convez-ai shulain/MobausStudio

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
1
Saúde
50%
Último commit
há 3 meses
Versão
v0.8.8
Issues abertas
0
Contribuidores
2

Casos de uso

  • Centralizar acesso a múltiplos modelos de IA em uma interface
  • Criar assistentes customizados com system prompts e permissões de ferramentas
  • Integrar IA a arquivos, bancos de dados e APIs via MCP
  • Colaboração multi-agente com discussões em roundtable

Veredito

Projeto maduro e bem-estruturado que resolve o problema real de fragmentação entre provedores de IA. A arquitetura em Tauri + React é sólida, e a integração nativa com MCP servers abre possibilidades autênticas de orquestração. Recomendado para quem trabalha com múltiplos modelos ou precisa de agents customizados.

README

A tradução para o português deste README ainda está na fila.

MobausStudio

An open-source, cross-platform AI client with multi-model support, MCP extensions, and agent orchestration.

Release Downloads CI License Platform

中文 | Documentation | Changelog


Why MobausStudio?

Most AI clients lock you into a single provider. MobausStudio gives you one interface for 20+ AI providers, with OAuth magic login for your existing subscriptions (ChatGPT Plus, Claude Pro, GitHub Copilot, etc.) -- no API keys needed. Plus full MCP tool integration, multi-agent roundtable discussions, and 30,000+ built-in skills.


Features

Core

  • 20+ AI Providers -- OpenAI, Anthropic, Google Gemini, DeepSeek, Qwen, Kiro, Ollama, and more
  • 4 Protocols -- OpenAI Chat Completions, Anthropic Messages, Google Gemini, AWS Bedrock
  • OAuth Magic Login -- Use your ChatGPT Plus / Claude Pro / GitHub Copilot subscriptions directly; no API key needed
  • Custom Providers -- Add any OpenAI-compatible API endpoint

Advanced

  • MCP Extensions -- Connect AI to file systems, databases, APIs via Model Context Protocol
  • Agent System -- Create custom AI assistants with system prompts, skills, and tool permissions
  • Roundtable Meeting -- Multi-agent collaboration: assign roles (architect, QA, PM) to discuss a topic from different perspectives
  • Skills System -- 30,000+ prompt templates for translation, writing, coding, analysis, and more

Platform

  • Cross-Platform Desktop -- macOS, Windows, Linux native apps via Tauri 2; this is the full production AI client
  • Web Static Preview -- Self-hostable static build for UI exploration and local configuration preview
  • Docker Static Preview -- One-command containerized static preview
  • Auto Update -- Built-in updater with signed releases
  • i18n -- English and Chinese UI

Tech Stack

LayerTechnology
DesktopTauri 2 (Rust)
FrontendReact 19 + TypeScript 5.8
StylingTailwind CSS 4
BuildVite 7
TestingVitest + Testing Library (Frontend), cargo test (Rust)
DocsVitePress

Download & Install

Desktop Application

Download from the Releases page:

PlatformFile
macOS (Apple Silicon).dmg (aarch64)
macOS (Intel).dmg (x64)
Windows.exe or .msi
Linux (Debian/Ubuntu).deb
Linux (Fedora/RHEL).rpm
Linux (Universal).AppImage

Docker

Important: The Docker image serves the static Web preview only. It does not include the Tauri/Rust backend, so AI chat requests, OAuth magic login callbacks, MCP execution, filesystem-backed secure storage, and auto-updates require the desktop app. The official GHCR image is published for linux/amd64 and linux/arm64.

docker run -d -p 8080:80 --name mobaus-studio ghcr.io/shulain/mobausstudio:latest
# Visit http://localhost:8080
services:
  mobaus-studio:
    image: ghcr.io/shulain/mobausstudio:latest
    ports:
      - "8080:80"
    restart: unless-stopped

Web Static Preview (Self-Hosted)

The self-hosted Web build is a static preview. Use a desktop release for the complete production AI-client experience.

Download MobausStudio-web.zip from Releases, extract and serve:

unzip MobausStudio-web.zip -d mobaus-web
cd mobaus-web && npx serve .

Development

Prerequisites

Quick Start

# Install dependencies
npm install

# Start frontend dev server
npm run dev

# Start full Tauri app (frontend + Rust backend)
npm run tauri dev

Build

# Web/static preview build
npm run build

# Desktop build
npm run tauri build

Test

# Frontend tests
npm test

# Rust tests
cd src-tauri && cargo test

# Coverage
npm run test:coverage

Release

Note: Version numbers in package.json and tauri.conf.json are placeholders (0.0.0-dev). CI sets the version automatically from git tags.

# Tag and push to trigger a release build
git tag v1.0.0
git push origin v1.0.0

Project Structure

MobausStudio/
├── src/                    # React frontend
│   ├── components/         # UI components (common, features, layout)
│   ├── services/           # Business logic (auth, providers, MCP, models)
│   ├── hooks/              # React hooks
│   ├── i18n/               # Internationalization (en, zh)
│   └── test/               # Frontend tests
├── src-tauri/              # Rust backend
│   └── src/
│       ├── lib.rs          # Tauri commands & core logic
│       ├── protocol/       # AI protocol implementations
│       ├── mcp/            # MCP client (stdio + HTTP transport)
│       └── services/       # Backend services (config exporter, etc.)
├── docs/                   # Internal dev documentation
├── user-guide/             # User documentation (VitePress, EN + ZH)
└── .github/workflows/      # CI/CD (ci, release, docs)

System Requirements

PlatformMinimum
macOS10.15 (Catalina) or later
WindowsWindows 10 (1803) or later
Linuxglibc 2.31+ (Ubuntu 20.04+)

Contributing

Contributions are welcome! Please:

  1. Fork the repository
  2. Create a feature branch (git checkout -b feat/amazing-feature)
  3. Commit your changes
  4. Push to the branch (git push origin feat/amazing-feature)
  5. Open a Pull Request

Feedback

Found a bug or have a suggestion? Open an issue.

Please include: OS & version, MobausStudio version, steps to reproduce, and any relevant screenshots or logs.


License

MIT License


Links

Ver o original no GitHub

Bom para

Relacionados