TagUI

NOTICE: AI Singapore has discontinued maintenance and support of TagUI, but the project continues to be supported by open-source contributors and the community.

Check out this YouTube learning series. Take the free course. Ask any questions at our Telegram.

Download v6.114 | Usage Guide | Demos | Samples | Slides | Podcast | Video | 中文

---

Write flows in simple TagUI language and automate away repetitive time-consuming tasks on your computer. Tasks include those on websites (native support for Chrome and Edge), desktop apps, or the command line. The TagUI project is open-source and free forever. It's easy to setup and use, and works on Windows, macOS and Linux.

Besides English, flows can be written in 22 other languages, so you can do RPA using your native language. Check out this demo video automating data collection in 4 different languages. With the new TagUI turbo mode, you can even run your automation 10X faster than normal human speed!

Language designed for RPA

In TagUI language, you use steps like click and type to interact with identifiers, which include web identifiers, image snapshots, screen coordinates, or even text using OCR. Below is an example to login to Xero accounting:

https://login.xero.com/identity/user/login
type email as [email protected]
type password as 12345678
click Log in

// besides web identifiers, images of UI elements can be used
type email_box.png as [email protected]
type password_box.png as 12345678
click login_button.png

Grabbing data from a table on a website can be as easy as below (where n is the nth table on the webpage)

table n to forex_rates.csv

Moving data between TagUI and Excel is as easy as using standard Excel formula that you are familiar with

top_salesman = [Monthly Report.xlsx]August!E11

Do RPA Any Way You Want

You can use TagUI MS Office Plug-ins (sample doc) to easily create and deploy Word doc as RPA robots, and set up RPA data parameters using Excel. Enjoy a full-featured RPA IDE with toolbar of TagUI steps and tooltips, snapshot tool to automate using computer vision, task pane for settings and run output.

You can also create and edit your RPA robots using commonly used text editors like Notepad, Notepad++, VS Code, Sublime, TextEdit, Vim, etc. For VS Code users, you can install TagUI language extension. For Notepad++ users, you can download TagUI plug-ins here for syntax highlighting, shortcuts and snippets.

Some users might find it interesting or useful to use AI and machine learning (in particular LLM large language models), to help generate a template script, then they make the fine-tuning accordingly. See this issue on some questions that I asked Anthropic's Claude 3.5 Sonnet and its responses.

Ecosystem and Communities

TagUI has a bustling user community, and extended community champions create new RPA tools for their own communities, based on TagUI. Python users can pip install rpa to use the #1 Python RPA package. There is also TagUI for C# .NET. For event-driven RPA with thousands of connectors, do check out TagUI module for Node-RED, a popular free and open-source workflow automation tool.

For Microsoft Power Automate Desktop users, you'll be happy to know that there's 2-way integration with TagUI out of the box (for business continuity if you switch between the 2 apps). Also, do check out other leading open-source RPA tools, to see if they meet your needs better - OpenRPA & OpenFlow, Robocorp. All of them support enterprise-grade orchestrating and running TagUI robots from their orchestrator.

Enterprise Security by design

Security Considerations

• TagUI default implementation is an on-user-computer on-prem RPA tool that does not exist on any cloud
• TagUI is not a SaaS or software on the cloud running on vendor's cloud, it runs on actual users' computers
• Industry-specific certifications like PCI-DSS, HIPAA, SOX aren't applicable because TagUI doesn't store data
• In decentralised bottom-up RPA, not advisable and no need for bot credentials as users are held accountable

Data Considerations

• For data at rest, storage encryption would be on user's computer's OS-level as it is run on user's computer
• For data in use, recommend user to manually enter sensitive info like password before letting robot take over
• For data in motion, users' enterprise app websites are now https by default for secure data entry and retrieval

More Information

• See this guide on enterprise installation, including whitelisting details, TagUI architecture and dependencies
• With -report option, there is a summary and detailed logs of robots, with support for centralised reporting

How to get started

Join the community and ask any questions at our Telegram chat group. Watch this YouTube learning series. Take TagUI free course over one morning or afternoon, and start using the most popular open-source RPA software. Share this TagUI slide deck with your team or customers to win their buy-in to use TagUI.

If you are maintaining your own fork of TagUI (for eg tech leads, RPA consultants, individual developers), see this maintainer training video series to understand how TagUI works behind the scenes, and for you to modify all aspects of TagUI and extend the software to your customers' exact needs, made-to-measure.

TagUI v5.11

For technical details of TagUI, such as architecture diagram and codebase structure, see the old homepage

Credits

Open-source project	Maintainer	From	How does this contribute to TagUI project
TagUI for China	报表哥	China	usage guide and repository in Chinese
TagUI for Notepad++	Md Ardyansyah	Indonesia	various TagUI plug-ins for Notepad++
TagUI for VS Code	Subhas Malik	India	language extension for Visual Studio Code
TagUI for Robocorp	Nived N	India	run TagUI in Robocorp or Robot Framework
TagUI for Node-RED	Allan Zimmermann	Denmark	low-code event-driven workflow automation
TagUI for C# .NET	Allan Zimmermann	Denmark	C# version of TagUI (Install-Package tagui)
TagUI for Docker	Allan Zimmermann	Denmark	replicable RPA environment for everyone
RPA for Python	Ken Soh	Singapore	Python version of TagUI (pip install rpa)
TagUI v3.0	Ken Soh	Singapore	personal project before AI Singapore
SikuliX	Raimund Hocke	Germany	computer vision, OCR, input hardware
CasperJS	Nicolas Perriault	France	high-level JavaScript execution engine
PhantomJS	Ariya Hidayat	Indonesia	foundation JavaScript execution engine
SlimerJS	Laurent Jouanneau	France	browser automation for FireFox <= v59

Sponsor

This project was previously supported by the National Research Foundation, Singapore under its AI Singapore Programme (AISG-RP-2019-050). Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of National Research Foundation, Singapore.