RPA

Name: RPA
Author: A9T9

Verified

Browser and desktop automation with an MCP server for AI assistants. Record macros or write JavaScript to automate websites, extract data, and control desktop apps with OCR and image recognition.

2,027stars

411forks

TypeScript

Installation

# Add to your Claude Code skills
git clone https://github.com/A9T9/RPA

Getting Started

Guides for using ai agents skills like RPA.

Caveman: Cut Claude Token Use by 65%
How agent-side prompt compression works, when to use it, and when not to.
What is an AI Skills Marketplace?
Definitions, how marketplaces work, and how to choose between them in 2026.
Getting Started with AI Skills

Security ReportVerified

Last scanned: 9/22/2026

{
  "issues": [],
  "status": "PASSED",
  "scannedAt": "2026-09-22T09:02:34.287Z",
  "npmAuditRan": false,
  "pipAuditRan": true,
  "promptInjectionRan": true
}

Compare with similar skills

See how RPA compares with popular alternatives.

RPA vs ecc

See comparison

RPA vs hermes agent

See comparison

RPA vs n8n

See comparison

README.md

Frequently Asked Questions

What is RPA?

RPA is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by A9T9. Browser and desktop automation with an MCP server for AI assistants. Record macros or write JavaScript to automate websites, extract data, and control desktop apps with OCR and image recognition. It has 2,027 GitHub stars.

Is RPA safe to use?

Yes. RPA passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install RPA?

Clone the repository with "git clone https://github.com/A9T9/RPA" and add it to your Claude Code skills directory (see the Installation section above).

What programming language is RPA written in?

RPA is primarily written in TypeScript. It is open-source under A9T9 on GitHub, so you can review or fork the full source.

Are there alternatives to RPA?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh RPA against similar tools.

Agentic AI for Beginners

Build your first AI agent from scratch - tool use, ReAct pattern, memory, deployment

41 minBeginner

Comments (0)

to leave a comment.

No comments yet. Be the first to share your thoughts!

Related Skills

ECC

by affaan-m

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

264,926

x64dbg-mcp-server boss-agent-cli

Ui.Vision - Browser and Desktop Automation

Automate websites and desktop applications with macros, JavaScript, or an AI assistant. Ui.Vision combines browser automation, OCR and image recognition, with an MCP server that lets AI agents create, edit and run reusable macros in your browser.

Install the extension · Connect an AI assistant with MCP · API reference · User forum

What can you automate?

Browser tasks: fill forms, extract web data, download reports and repeat workflows in your existing browser session.
Website testing: record and replay interactions, use Selenium IDE commands, and run checks with different input data.
Visual tasks: find on-screen text with OCR and locate controls by image when HTML selectors are unavailable.
Desktop workflows: automate applications and remote desktop interfaces using visual recognition and mouse and keyboard input. Desktop automation requires the additional Ui.Vision desktop components (XModules).
AI-assisted automation: ask the built-in AI assistant or an external MCP client to write and run macros, then inspect the resulting script, logs and screenshots.

Use the macro recorder and command table, write JavaScript with the uiv.* API, or let an AI assistant author the macro. The resulting automation can be saved and run again.

Get started

Install the browser extension; you do not need to build this repository:

Open Ui.Vision and record a browser task, or start with a macro from the extension's demos. For desktop input and additional native capabilities, see the XModules installation guide.

The browser extension is open source and free for personal and commercial use. See the Ui.Vision website for desktop components and available editions.

Connect an AI assistant with MCP

Ui.Vision provides a Model Context Protocol (MCP) server through the uivision-mcp-bridge package. It connects MCP clients such as Claude Code, Claude Desktop and Cursor to the Ui.Vision browser extension.

An agent can inspect a page, create or edit a macro, run it, and use logs and screenshots to check the result. The automation executes through Ui.Vision in your browser.

Start with the MCP setup guide. The installer command is:

npx uivision-mcp-bridge --setup

The installer registers Ui.Vision with supported MCP clients found on your machine. Then enable the MCP bridge in Ui.Vision → Settings → AI, complete the connection steps in the guide, and restart your MCP client. Node.js and npm are required to run the installer.

For a first task, ask your connected assistant:

Use Ui.Vision to open https://example.com, read the page heading, and save a reusable macro for this task.

See mcp/README.md for bridge source, configuration and troubleshooting.

JavaScript automation and documentation for AI assistants

The uiv.* JavaScript API covers page elements, visual matching, OCR, browser and desktop input, screenshots, tabs, CSV files and downloads. Ui.Vision macros execute sequentially: write API calls without async or await.

When asking an AI assistant to write a macro, give it the API reference below so it uses Ui.Vision's supported methods and runtime conventions.

Ui.Vision JavaScript API and AI authoring guide: API syntax and automation recipes used by the built-in assistant.
Plain-text reference for AI assistants: documentation that can be supplied directly as context.
Classic command-to-JavaScript mapping: uiv.* equivalents and the uiv.run('command', 'target', 'value') bridge for classic commands.
Selenium IDE command reference: documentation for command-table macros.

Help and support

Ask questions and share automation examples in the Ui.Vision user forum, where users, support staff and developers participate.

Build from source

Building the extension is not required if you "only" want to use it.

You can install UI.Vision directly from the Chrome, Edge or Firefox stores, which is the easiest and the recommended way of using the Ui.Vision browser extension. Older versions can be found in the RPA software archive.

The information below is only required and intended for developers:

The project uses Node V20.11.1 and NPM V10.2.4

If you have any questions, please contact us at TEAM AT UI.VISION - Thanks!

Build the extension bundle

npm i -f

npm run build

npm run build-ff

npm run build creates the Chrome/Edge build in dist, npm run build-ff creates the Firefox build in dist_ff.

Develop

npm i -f

npm start

Use npm run start-ff for the Firefox variant. Both run webpack in watch mode, so the bundles in dist (Chrome/Edge) and dist_ff (Firefox) are rebuilt on every change.

Once done, the ready-to-use extension code appears in the /dist directory (Chrome, Edge) or /dist_ff directory (Firefox). Load it via chrome://extensions → "Load unpacked" (Chrome/Edge) or about:debugging → "Load Temporary Add-on" (Firefox).

Repository layout

src/ - the extension source (React UI, side panel, macro player, commands)
extension/ - static extension assets and manifest.json (Manifest V3)
mcp/ - the Ui.Vision MCP bridge, which lets Claude Code and other MCP clients create, edit and run macros. See mcp/README.md
dist/, dist_ff/ - build output for Chrome/Edge and Firefox