Your First Automation Project

This guide walks you through creating your first automation project from scratch. By the end, you will have a working automation that visits a website and extracts data. There are four different methods to create a project — choose whichever feels most natural to you.

Method 1: Ask the AI Assistant (Easiest)

The fastest way to create a project is to simply describe what you want in plain English. The AI Assistant will analyze the target website, generate the project configuration, and create it for you.

Steps

1. Click AI Assistant in the sidebar (the home icon at the top).
2. You will see the AI chat interface. If this is your first time, you may see a welcome screen with quick-start options.
3. Type your request in the text box at the bottom. For example:

“Go to https://news.ycombinator.com and extract the title, link, and points of each story on the front page.”

4. Press Enter or click the send button.
5. The AI will begin working. You will see a thinking indicator while it processes your request.
6. The AI analyzes the target page using its built-in tools. It may show you progress messages like “Launching stealth browser to analyze the page…” and “Analyzing page structure…”
7. Once analysis is complete, the AI generates a project configuration containing commands and datasets.
8. The generated project appears in the chat as a structured block.
9. Click Run Project to execute it immediately, or navigate to Automation Projects to review and edit it first.

Tips for Best Results

• Be specific about what data you want (e.g., “product name, price, and image URL” rather than just “data”).
• Include the full URL of the target website.
• If the site requires login, mention that (e.g., “Log in with username and password, then extract…”).
• If the AI asks for clarification (such as login credentials), respond in the chat and it will continue building the project.

What Happens Behind the Scenes

When you ask the AI to build a project, it:

1. Navigates to the target URL using a stealth browser.
2. Takes a full-page screenshot and analyzes the page structure.
3. Identifies interactive elements, data patterns, and forms.
4. Generates a project configuration with commands (like goto, extract_data) and datasets (defining what to extract and how).
5. Returns the complete project for you to run.

---

Method 2: Record Your Actions

If you prefer to show RTILA what to do instead of describing it, use the Record feature. RTILA watches your browser actions and converts them into automation commands.

Steps

1. Click Automation Projects in the sidebar.
2. Click Record in the ribbon toolbar at the top.
3. A new project is created and the Visual Builder browser opens immediately with recording active.
4. You will see a red recording indicator bar at the top of the browser that says “Recording Actions… (Alt+Shift+X to stop)”.
5. Perform your actions normally:
   • Click buttons and links
   • Fill in form fields
   • Select dropdown options
   • Navigate between pages
   • Type text into inputs
6. RTILA records every action as a command.
7. When you are finished, press Alt+Shift+X (Windows/Linux) or Option+Shift+X (Mac), or click the Stop button on the recording indicator.
8. A summary appears showing how many commands were recorded.
9. Go back to RTILA X to review your recorded commands in the project.

What Gets Recorded

Your Action	Recorded Command
Clicking an element	click
Double-clicking	click (with click_count: 2)
Right-clicking	click (with button: “right”)
Filling a form field	fill
Selecting a dropdown option	select_option
Pressing Enter, Tab, etc.	press
Hovering over an element (0.5+ seconds)	hover
Scrolling the page	scroll
Navigating via address bar	goto
Clicking a link that loads a new page	wait_for_load_state
Switching browser tabs	switch_to_tab
Resizing the browser window	set_viewport_size
Dragging and dropping	drag_to
Selecting text	select_text

Recording Tips

• Click precisely on the element you want to interact with.
• Wait for pages to load before clicking the next element.
• Use the address bar for navigation — it records a clean goto command.
• Avoid rapid clicking — give elements time to respond.
• Hover intentionally — hovering is only recorded if you pause on an element for at least 0.5 seconds.

Loop Recording

While recording, you can click Start Loop on the recording indicator to begin a loop recording session. Perform the actions you want to repeat, then click Stop Loop. A configuration dialog appears where you choose:

• Repeat fixed number of times — runs the loop a specific number of times.
• While element is visible — keeps looping as long as a specific element exists.
• For each item in list/dataset — iterates over a list variable or extracted dataset.

---

Method 3: Build Manually

For full control, build your automation command by command in the Visual Builder.

Steps

1. Click Automation Projects in the sidebar.
2. Click New in the ribbon toolbar.
3. A new project is created and the Settings modal opens automatically.
4. Enter a Project Name (RTILA generates a fun spider-themed name by default, like “Agile Tarantula - Data Extractor”).
5. In the Starting URLs field, enter the website URL you want to automate (one per line).
6. Configure any other settings you need (see Projects for details).
7. Click Save & Launch Visual Builder.
8. The browser opens with the RTILA side panel on the right.
9. In the side panel, switch to the Web Automation tab.
10. Click Add at the bottom to add commands one by one.
11. A command picker appears. Search or browse for the command you need (e.g., goto, click, extract_data).
12. Click a command to add it to your project.
13. Configure each command’s parameters in the panel that appears.

Adding Data Extraction

1. In the side panel, switch to the Data Extraction tab.
2. Click + to add a new dataset.
3. Use the element picker (the pointer icon) to click on elements on the page and define what to extract.
4. Add properties (columns) to your dataset by clicking + in the properties area.
5. For each property, choose the selector, type (text, attribute, etc.), and any transformations.
6. The live preview table at the bottom shows you what data will be extracted.

---

Method 4: Use a Template

RTILA X includes pre-built templates for common automation tasks.

Steps

1. On the AI Assistant welcome screen, click Explore Templates.
2. A template browser opens showing available templates organized by platform.
3. Browse the templates or use the search bar to find one that matches your needs.
4. Click on a template to select it.
5. The template is imported as a new project in your workspace.
6. You can then edit the project settings, commands, and datasets to customize it for your specific needs.

---

Running Your Project

Once your project is built (by any method), you can run it.

Steps

1. Go to Automation Projects in the sidebar.
2. Click on your project in the table to select it.
3. Click Run in the ribbon toolbar.
4. The Live Monitor modal opens, showing real-time progress.

Live Monitor

The Live Monitor has three tabs:

Magic Dashboard

The Magic Dashboard shows three status cards:

Card	What It Shows
Bot Evasion Status	Whether humanoid mouse movements are active. Shows “Active: Humanoid movements engaged” when stealth mode is working.
Resilience Engine	Whether checkpoints are being saved. Shows “Saving Checkpoint…” when auto-saving progress.
AI Self-Correction	The number of errors the AI has automatically fixed during this run.

Live Logs

The Live Logs tab shows a scrolling feed of every action the automation is performing. Each log entry shows:

• A timestamp
• The log level (INFO, WARN, ERROR, etc.)
• A description of what is happening

You can toggle Hide Technical Jargon to filter out low-level technical messages and see only the important milestones.

Live Memory

The Live Memory tab shows all variables currently stored in the automation’s memory. This is useful for debugging — you can see what values have been extracted and stored as the automation runs.

Stopping a Running Project

1. While the Live Monitor is open, click Stop Project.
2. A confirmation dialog appears.
3. Click Stop Project again to confirm.
4. The automation stops gracefully.

---

Viewing Your Results

After your project finishes running, you can view the extracted data.

Steps

1. Go to Scraped Data in the sidebar.
2. Use the dropdown filters at the top:
   • Select Project — choose your project.
   • Select Run — choose the specific run (session) you want to view.
   • Select Dataset — choose which dataset to display.
3. Your extracted data appears in a table.
4. Click Export to download the data.

Export Formats

Format	Extension	Best For
CSV	.csv	Opening in Excel or Google Sheets
JSON	.json	Developers and API integrations
XLSX	.xlsx	Excel spreadsheets
Parquet	.parquet	Large datasets and data analytics
SQL	.sql	Importing directly into a database

Export Steps

1. Click Export on the Scraped Data page.
2. Choose your desired format.
3. For CSV, you can choose to overwrite or append to an existing file.
4. Click Save and choose where to save the file on your computer.

---

What’s Next?

Now that you have created and run your first project, explore these topics:

• Commands Reference — Learn all available commands
• Datasets Guide — Master data extraction
• Stealth Features — Bypass anti-bot protection
• Triggers — Send data to databases, cloud storage, and more
• Task Scheduler — Run automations automatically on a schedule