Python and Robot Framework libraries for Robotic Process Automation, covering browsers, Excel, email, PDF, cloud APIs and desktop apps.
A9T9/RPA
Browser and desktop automation extension with a macro recorder, JavaScript API, OCR, image recognition, and an MCP server for AI assistants.
Open-source alternative to:UiPath
Automation Anywhere
- Stars
- 2,101
- Forks
- 422
- Open issues
- 10
- Last commit
- 3 days ago
What is it?
Ui.Vision is a browser extension for automating websites and desktop applications. Users can record a macro, edit it as a command table, or write JavaScript with the uiv.* API. It covers form filling, web data extraction, file downloads, and repeated workflows that run in the user's existing browser session.
For pages without usable HTML selectors, Ui.Vision can find on-screen text with OCR and locate controls by image matching. Desktop automation uses visual recognition together with mouse and keyboard input, and requires the separate Ui.Vision desktop components (XModules). Selenium IDE commands are supported for test-style macros.
The project includes an MCP server, provided through the uivision-mcp-bridge package, that connects MCP clients such as Claude Code, Claude Desktop and Cursor to the extension. An AI agent can inspect a page, create or edit a macro, run it, and review logs and screenshots. The repository also contains the extension source, built with React and webpack for Chrome, Edge and Firefox.
Key features
- 01Records browser actions as macros or edits them as a command table.
- 02Automates websites with the uiv.* JavaScript API.
- 03Finds on-screen text with OCR and locates controls by image.
- 04Provides an MCP server so AI assistants can create, edit and run macros.
- 05Supports Selenium IDE commands for test-style macros.
- 06Builds for Chrome, Edge and Firefox.
Use cases
- Filling out repetitive web forms or downloading reports on a schedule.
- Extracting data from websites that lack a usable API.
- Running the same web test with different input data.
- Letting an AI agent draft and execute browser automation tasks.
- Automating desktop or remote desktop interfaces that have no HTML structure.
Quick start
- 1
Install the browser extension from the Chrome Web Store, Edge Add-ons or Firefox Add-ons
- 2
Connect an MCP client (optional) by running the installer
npx uivision-mcp-bridge --setup
Things to know
- Desktop automation requires separate XModules desktop components.
- The MCP bridge needs Node.js and npm installed.
- Building from source requires Node V20.11.1 and NPM V10.2.4.
In collections
Topics
FAQ
- What is RPA?
- Browser and desktop automation extension with a macro recorder, JavaScript API, OCR, image recognition, and an MCP server for AI assistants. Ui.Vision is a browser extension for automating websites and desktop applications. Users can record a macro, edit it as a command table, or write JavaScript with the uiv.* API. It covers form filling, web data extraction, file downloads, and repeated workflows that run in the user's existing browser session.
- Is RPA free to use? What license does it have?
- RPA is open source under the Other license. Check the license terms before using it in commercial products.
- Is RPA still maintained?
- Actively maintained. The last commit to RPA was on Nov 4, 2026.
- What can I use RPA for?
- Filling out repetitive web forms or downloading reports on a schedule. Extracting data from websites that lack a usable API. Running the same web test with different input data. Letting an AI agent draft and execute browser automation tasks. Automating desktop or remote desktop interfaces that have no HTML structure.
- How do I install RPA?
- Install the browser extension from the Chrome Web Store, Edge Add-ons or Firefox Add-ons Connect an MCP client (optional) by running the installer: npx uivision-mcp-bridge --setup
- Where can I get RPA?
- The source code and installation instructions are on GitHub: https://github.com/A9T9/RPA
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source execution layer that records a task once and compiles it into a local program agents or people can rerun with no model calls, verifying each save.
Free, open-source RPA tool that automates websites, desktop apps and command-line tasks with simple TagUI flow scripts.
Open-source browser automation that uses LLMs and computer vision to run web workflows on any site, extending Playwright with AI commands.
Records browser interactions once and replays them as deterministic workflows, falling back to Browser Use AI when a step fails.
MCP server that lets AI agents browse and inspect Firefox through WebDriver BiDi, for agentic browsing and developer-tool debugging.