Open Source · Apache 2.0

Your Desktop, on Autopilot

A personal AI agent that lives on it's own Mac, chats like humans through your daily chat apps, works like a human through commands, browser automation and desktop control.

Get Started View on GitHub

See It in Action

Real conversations. Real tasks. Fully autonomous.

Polymarket trading via Telegram
Telegram Autonomous prediction market trading
Portfolio management via Telegram
Telegram Portfolio tracking & order placement
Meeting scheduling via WeChat
WeChat Coordinating meetings & calendar invites
Video summary via WeChat
WeChat Extracting video content on demand
Multi-agent collaboration via WeChat
WeChat Coordinating tasks with other AI assistants
Code review via Lark
Lark Reviewing PRs & explaining code changes
Social posting via WeChat
WeChat Posting to social feeds autonomously
GitHub PR review
GitHub Commenting on PRs & running checks
Meeting scheduling via Gmail
Gmail Scheduling meetings and sending invites by email

What VisionClaw Can Do

An always-on agent with eyes and hands on computer.

👁

Vision & Screen Awareness

Sees screen just like a human — describes UI elements, reads images, and interprets documents to decide what to do next.

🖥

Desktop Control

Controls mouse and keyboard, operate any desktop apps, browsers, social apps, etc. and runs terminal commands.

🔌

Agent Ecosystem

Connects to any MCP servers and installs Skills on the fly to expand its capabilities without limits.

📅

Calendar & Email

Has its own Gmail account for email and Google Calendar for scheduling recurring tasks.

🧠

Self-Improving

Can learn new skills, create/install them at runtime, and upgrade itself to the latest version.

🗃

Persistent Memory

Remembers context across sessions with SQLite-backed memory, so it never forgets your preferences or past tasks.

Get Started

1

Prerequisites

Node.js >= 24.12.0, pnpm, macOS, an Anthropic API key, and a dedicated Gmail account for the agent.

2

Install & Build

Clone the repo, install dependencies, and build.

3

Run

The first run starts an interactive onboarding wizard that configures your API keys and channels.

git clone https://github.com/babelcloud/visionclaw.git
cd visionclaw
pnpm install
pnpm build
pnpm start