HomeProjectsArticlesPK BoardAbout
Explore Projects

UI-TARS Desktop

Active
GitHub TypeScript Apache-2.0

Description

ByteDance's open-source multimodal AI agent stack connecting cutting-edge AI models with agent infrastructure for GUI automation and computer control.

Key Features

  • Multimodal AI agent stack — Agent TARS + UI-TARS Desktop dual-project architecture covering CLI, Web UI, and desktop
  • GUI automation — Native desktop and browser control powered by the UI-TARS model
  • MCP tool integration — Seamless connection to external MCP servers for tool extensibility
  • Remote computer/browser control — One-click remote control of any computer or browser with zero config
  • Local/remote operator — Supports both local model and remote model operation modes
  • Sandbox execution — Dedicated AIO Agent Sandbox for isolated tool execution environments

Use Cases

💡 Automate web interaction tasks like booking, form filling, and data collection
💡 Cross-application GUI automation testing and quality assurance
💡 Control desktop applications via natural language instructions for complex workflows
💡 Build multimodal AI application prototypes requiring visual understanding

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (38.8k stars)
  • • Permissive open-source license (Apache-2.0)

Tags

multimodal-agent gui-automation computer-use mcp typescript

Categories

🌐 Browser Agent 🤖 Agent Framework

Quick Start

# Install Agent TARS CLI
npm install -g @agent-tars/cli

# Launch Agent TARS
agent-tars
Visit GitHub Visit Website

Related Projects

OpenAgent

5.6k · Go
Active A

Next-generation personal AI assistant powered by LLM, RAG and agent loops, supporting computer-use, browser-use and coding agent capabilities.

personal-assistantbrowser-agentcomputer-use +4
  • · Support for 30+ model providers, switchable per conversation without code changes
  • · Autonomous agent loops: browser automation, web search, shell execution and Office automation
  • · MCP-compatible integration with any MCP server (SSE/Stdio/StreamableHTTP)
View full details → View on GitHub

AIPex

1.2k · TypeScript
Active A

AI browser automation assistant as a Chrome extension, privacy-first with MCP support, alternative to Claude Chrome and Manus Browser Operator

browser-automationchrome-extensionmcp +2
  • · Zero-migration browser automation — runs in your existing Chrome/Edge browser, no new app to install
  • · Privacy-first BYOK model — your data never leaves your machine, bring your own API key
  • · MCP protocol support — 30+ browser automation tools accessible via Model Context Protocol
View on GitHub

Chrome DevTools MCP

50.6k · TypeScript
Active A+

MCP server providing Chrome DevTools capabilities to coding agents, enabling web debugging, performance analysis, and DOM manipulation automation.

mcpchrome-devtoolsbrowser-debug +2
  • · Performance Insights — Record Chrome performance traces and extract actionable optimization recommendations
  • · Advanced Browser Debugging — Analyze network requests, take screenshots, check console with source-mapped stack traces
  • · Reliable Automation — Puppeteer-based browser automation with automatic wait for action results
View full details → View on GitHub

VoltAgent

10.5k · TypeScript
Active A+

VoltAgent is an end-to-end AI Agent engineering platform combining a TypeScript framework with the VoltOps console for workflow, MCP, memory, and observability.

workflowagent-frameworktypescript +2
  • · Core agent runtime - typed roles, tools, memory, and model providers wired in one place
  • · Declarative workflow engine - multi-step automations described in code, not graph editors
  • · Supervisor and sub-agent topology - specialized agents coordinated by a supervisor runtime
View full details → View on GitHub