AIPex
AIPex: AI browser automation assistant, no migration and privacy first. Alternative to Manus Browser Operator、 Claude Chrome and Agent Browser
claude mcp add --transport stdio aipexstudio-aipex npx -y aipex-mcp-bridge
How to use
AIPex exposes an MCP bridge that lets AI coding agents control the AIPex browser automation extension directly via the MCP protocol. The bridge runs in your environment (via npx) and exposes a WebSocket interface that agents connect to through MCP. Once connected, the agent can issue tool-capability commands that map to AIPex browser actions (such as tab management, navigation, text input, and other browser automation primitives) using a standardized MCP format. The extension receives these commands and executes them inside the browser, enabling automated tasks powered by natural language or structured agent plans. You’ll typically configure your agent to start the aipex-mcp-bridge with npx -y aipex-mcp-bridge, then point your agent’s MCP client at the bridge and connect via WebSocket to begin issuing commands. The setup supports multiple agent ecosystems (Cursor, Claude Code, VS Code Copilot, etc.).
How to install
Prerequisites:
- Node.js and npm (or npx available)
- Access to install npm packages from the command line
Step 1: Install Node.js (if not already installed)
- Download and install from https://nodejs.org/
Step 2: Install or run the MCP bridge Option A (recommended): Run via npx (no local installation required)
npx -y aipex-mcp-bridge
Option B (local install): Install the package locally and run from node_modules
npm install -D aipex-mcp-bridge
npx aipex-mcp-bridge
Step 3: Connect your MCP client
- Ensure the bridge is listening (default WebSocket port 9223)
- In your MCP client configuration, point to ws://localhost:9223
Step 4: Optional configuration
- If you want to customize environment variables or ports, refer to the bridge’s documentation and set them as needed in your environment prior to starting the bridge.
Additional notes
Tips and notes:
- The MCP bridge is expected to run with the AIPex extension installed in your browser; the extension exposes the actual browser automation capabilities.
- Default WebSocket URL for client connections is ws://localhost:9223. If you change the port, update your MCP client accordingly.
- Ensure your network firewall allows WebSocket traffic to the bridge port.
- You can run multiple MCP-enabled agents against the AIPex bridge by naming distinct MCP server entries in your configuration.
- The bridge supports 30+ browser tools and patterns as described in the AIPex skill documentation; consult mcp-bridge/README.md for advanced options and tool schemas.
Related MCP Servers
web-agent-protocol
🌐Web Agent Protocol (WAP) - Record and replay user interactions in the browser with MCP support
flowlens
FlowLens is an open-source MCP server that gives your coding agent (Claude Code, Cursor, Copilot, Codex) full browser context for in-depth debugging and regression testing.
coplay-unity-plugin
Unity plugin for Coplay
mcp-image
MCP server for AI image generation and editing with automatic prompt optimization and quality presets (fast/balanced/quality). Powered by Gemini (Nano Banana 2 & Pro).
adal-cli
The self-evolving coding agent that learns from your entire team and codebase. Less syncing. Less waiting. Deliver at the speed of thought.
izan.io
Turn Any Browser Action & Data Extraction into an AI Tool in 60 Seconds