RPA

RPA

An MCP server for browser and desktop automation, supporting macro recording, JavaScript scripting, and OCR image recognition.

MCPOfficeOpen source
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 2.0k
Source
mcp-github

Overview

RPA is an MCP server combining browser automation, OCR, and image recognition, enabling AI assistants to create, edit, and run reusable macros. It supports web form filling, data extraction, website testing, visual tasks, and desktop application automation. Users can achieve automation via macro recording or writing JavaScript, and can connect external MCP clients for more complex operations. Suitable for individuals and enterprises needing to automate repetitive tasks.

Capabilities

  • ▪Web form filling and data extraction
  • ▪Website testing and interaction recording
  • ▪OCR text recognition
  • ▪Image recognition control
  • ▪Desktop application automation

Use cases

Automatically fill out and submit web formsRegularly download reports from websitesAutomate desktop software operationsUse OCR to recognize text on the screen

Setup

Requires: Node.jsnpmAPI Key
npx uivision-mcp-bridge --setup

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I get started with RPA?

Install the browser extension and record a browser task, or start from example macros.

Does it support desktop application automation?

Yes, but requires additional installation of the Ui.Vision desktop component (XModules).

Related skills