LinkedIn Automation

A sophisticated Go application for automating LinkedIn login and interactions with human-like behavior simulation using the Rod browser automation library.

Overview

This project uses Rod to automate browser interactions with LinkedIn. It features advanced stealth techniques including random mouse movements, natural scrolling patterns, and realistic typing speeds to avoid detection by anti-bot systems.

Features

• Automated LinkedIn Login - Logs into LinkedIn with email/password credentials
• Human-like Behavior Simulation:
  • Random mouse movements (3-5 movements with natural pauses)
  • Random page scrolling (3-5 scrolls with reading time)
  • Realistic typing speeds
  • Random delays between actions (simulating human reaction time)
• Stealth Fingerprint Masking - Masks automation indicators to avoid detection
• Comprehensive Logging - Detailed logs of all actions performed
• Browser Automation - Full control over browser interactions using Rod library
• Environment-based Credentials - Secure credential management via .env file

Prerequisites

• Go 1.24.5 or higher
• Chrome/Chromium browser (Rod will auto-download if not present)
• Windows Defender exclusion (optional, see troubleshooting)
• .env file with LinkedIn credentials

Installation

1. Clone the repository:

git clone https://github.com/YOUR_USERNAME/linkedin-automation.git
cd linkedin-automation

2. Install dependencies:

go mod tidy

3. Create a .env file in the project root:

LINKEDIN_EMAIL=your_email@example.com
LINKEDIN_PASSWORD=your_password

4. Build the project (optional):

go build

Quick Start (5 Minutes)

Prerequisites Check

# Verify Go is installed
go version  # Should be 1.24.5 or higher

# Verify you're in the project directory
cd linkedin-automation
ls -la  # Should show main.go, README.md, go.mod, etc.

Setup Steps

# Step 1: Download dependencies
go mod tidy

# Step 2: Create credentials file (.env)
# On Windows (PowerShell):
@"
LINKEDIN_EMAIL=your_email@example.com
LINKEDIN_PASSWORD=your_password
"@ | Out-File -Encoding UTF8 .env

# On Linux/Mac:
cat > .env << EOF
LINKEDIN_EMAIL=your_email@example.com
LINKEDIN_PASSWORD=your_password
EOF

# Step 3: Verify .env file exists
ls -la .env  # Should show .env file

# Step 4: Run the application
go run main.go

What to Expect (Timeline)

• 0-2 seconds: "Starting LinkedIn Automation" log appears
• 2-3 seconds: Browser window launches (Chrome opens)
• 3-5 seconds: Navigates to LinkedIn login page
• 5-15 seconds: Enters credentials and logs in (watch it type)
• 15-20 seconds: "Login Successful" message appears
• 20-25 seconds: Mouse movements visible on page (cursor moving)
• 25-30 seconds: Page scrolls up and down automatically
• 30+ seconds: Browser stays open showing home page

Usage

Run the application:

go run main.go

What the Application Does

1. Loads credentials from the .env file
2. Launches a Chrome browser in non-headless mode (visible)
3. Navigates to LinkedIn login page
4. Performs automated login with human-like delays
5. Verifies successful login by checking page URL
6. Applies fingerprint masking to avoid detection
7. Simulates human behavior:
   • Random mouse movements across the page
   • Random scrolling with natural pauses
   • Realistic timing between all actions
8. Keeps browser open for 30+ seconds showing the results

Expected Output

[2025-12-24 19:08:28] INFO: Starting LinkedIn Automation
[2025-12-24 19:08:29] INFO: Opening Linkedin Login page
[2025-12-24 19:08:30] INFO: Locating email input field
[2025-12-24 19:08:31] INFO: Locating password input field
[2025-12-24 19:08:32] INFO: Locating and clicking on sign in button
[2025-12-24 19:08:33] INFO: Checking if login was successful
[2025-12-24 19:08:34] INFO: Login Successful - Redirected to home page
[2025-12-24 19:08:35] INFO: Applying fingerprint masking...
[2025-12-24 19:08:36] INFO: Starting human-like behavior simulation...
[2025-12-24 19:08:37] INFO: Performing stealth actions - simulating human-like behavior
[2025-12-24 19:08:38] INFO: Executing random mouse movements...
[2025-12-24 19:08:39] INFO: Executing random page scrolling...
[2025-12-24 19:08:40] INFO: Stealth actions completed

First Run Checklist

Before running the application, verify:

• Go 1.24.5+ is installed: go version
• You're in the project directory: pwd or cd linkedin-automation
• Dependencies are installed: go mod tidy
• .env file exists with credentials: cat .env
• .env has valid LinkedIn email and password
• Chrome/Chromium is installed (or let Rod download it)
• No other applications are using port 9222 (Rod's default debugging port)
• Windows Defender isn't blocking Rod (add exclusion if needed)

Build & Run Commands

Development (with hot reload)

# Install go-task or use go run directly
go run main.go

Production Build

# Build executable
go build -o linkedin-automation.exe

# Run the executable
./linkedin-automation.exe

Build with Optimizations

# Smaller binary size
go build -ldflags="-s -w" -o linkedin-automation.exe

# Run it
./linkedin-automation.exe

Cross-Platform Building

# Build for Windows (if on Windows)
go build -o linkedin-automation.exe

# Build for Linux (if on Linux/Mac)
go build -o linkedin-automation

# Build for macOS (if on Mac)
go build -o linkedin-automation

# Cross-compile Windows binary from Linux/Mac
GOOS=windows GOARCH=amd64 go build -o linkedin-automation.exe

Verification Steps (Testing Actual Functionality)

Step 1: Verify Compilation

go build
# Should complete with no errors
ls -la linkedin-automation.exe  # File should exist

Step 2: Run and Verify Each Feature

A. Verify Credential Loading

go run main.go
# Check logs for: "Starting LinkedIn Automation"
# Check that browser launches

B. Verify LinkedIn Login

• Watch the browser window
• See email field being filled automatically
• See password field being filled automatically
• See the "Sign in" button being clicked
• Wait for page to load and check logs for "Login Successful"

C. Verify Mouse Movements ✅

• After login, watch the mouse cursor on the page
• You should see the cursor move to 3-5 different positions
• Each movement is followed by a 300-800ms pause
• Check logs for: "Executing random mouse movements..."

D. Verify Page Scrolling ✅

• After mouse movements, watch the page scroll
• Page should scroll down 3-5 times
• Each scroll is 200-600 pixels
• Followed by 800-1500ms pause (reading time)
• Check logs for: "Executing random page scrolling..."

E. Verify Logging

• Check that every action is logged
• Look for log format: [TIMESTAMP] INFO: Action name
• Verify no ERROR messages in logs

Step 3: Test with Different Credentials

# Edit .env with test account
# Run again
go run main.go

# Verify it works with different credentials

Step 4: Test Build Executable

# Build the project
go build

# Run the executable instead of 'go run'
./linkedin-automation.exe

# Should work identically to 'go run main.go'

Success Indicators ✅

• Browser launches automatically
• Credentials are entered without user interaction
• Login completes successfully
• Mouse movements are visible on the page
• Page scrolling occurs automatically
• All actions are logged with timestamps
• No error messages in console
• Program completes without crashing
• Browser closes cleanly when done

Project Structure

linkedin-automation/
├── main.go                    # Entry point - orchestrates the automation workflow
├── go.mod                     # Go module definition
├── go.sum                     # Dependency checksums
├── .env                       # Environment variables (credentials)
├── README.md                  # This file
│
├── internal/
│   ├── automation/
│   │   └── login.go           # LinkedIn login automation logic
│   │
│   ├── browser/
│   │   ├── browser.go         # Browser initialization, page opening, stealth actions
│   │   └── fingerprint.go     # Anti-detection fingerprint masking
│   │
│   ├── logger/
│   │   └── logger.go          # Centralized logging utility
│   │
│   ├── stealth/
│   │   ├── delay.go           # Random delay generation for human-like timing
│   │   ├── mouse.go           # Random mouse movement simulation (3-5 movements)
│   │   ├── scroll.go          # Random page scrolling simulation (3-5 scrolls)
│   │   └── typing.go          # Human-like typing speed simulation
│   │
│   ├── storage/
│   │   └── state.go           # Application state persistence
│   │
│   ├── config/                # Configuration management
│   └── models/                # Data models
│
└── pkg/
    ├── models/                # Public data models
    └── utils/                 # Public utility functions

tests/                         # Test files

Configuration

.env File Requirements

Create a .env file in the project root with your LinkedIn credentials:

LINKEDIN_EMAIL=your_email@example.com
LINKEDIN_PASSWORD=your_password

⚠️ Security Warning: Never commit .env files to version control. Add it to .gitignore.

Browser Settings

Edit internal/browser/browser.go to customize:

• Headless mode: Change Headless(false) to Headless(true) for invisible browsing
• Browser binary: Set custom Chrome path with .Bin("/path/to/chrome")
• Additional flags: Add launcher flags as needed

Stealth Settings

Customize behavior simulation in:

• internal/stealth/delay.go - Adjust delay ranges
• internal/stealth/mouse.go - Adjust movement ranges (lines 7-8)
• internal/stealth/scroll.go - Adjust scroll ranges (lines 7-8)
• internal/stealth/typing.go - Typing speed parameters

How Stealth Works

Mouse Movement Simulation

• Performs 3-5 random mouse movements
• Coordinates range: X(100-800px), Y(100-600px)
• Pause between movements: 300-800ms
• Avoids corners to appear natural

Scrolling Simulation

• Performs 3-5 random scroll actions
• Scroll distance per action: 200-600 pixels
• Pause between scrolls: 800-1500ms
• Mimics human reading behavior

Typing Simulation

• Realistic keystroke delays between characters
• Variable speed based on character type
• Occasional typos and corrections (optional)

Timing & Delays

• Login field detection: 800-1500ms delay
• Form submission: 1000-2000ms delay
• Page load verification: 3000-5000ms delay
• All delays are randomized to appear natural

Troubleshooting

Common Issues & Solutions

1. ".env file not found" error

Problem: Application exits with error about missing .env file

Solution:

# Verify .env exists in project root
ls -la .env

# If missing, create it
echo "LINKEDIN_EMAIL=your_email@example.com" > .env
echo "LINKEDIN_PASSWORD=your_password" >> .env

# Verify contents
cat .env

2. "go: command not found"

Problem: Go is not installed or not in PATH

Solution:

# Check if Go is installed
go version

# If not, download from https://golang.org/dl/
# Add Go to your PATH environment variable
# Restart PowerShell/Terminal after installation

3. "Failed to start Browser" error

Problem: Chrome/Chromium not installed

Solution:

# Option 1: Let Rod download Chrome automatically
# The first run will download a Chromium instance

# Option 2: Install Chrome manually
# Download from https://www.google.com/chrome/

# Option 3: Set custom Chrome path in browser.go
.Bin("C:\\Program Files\\Google\\Chrome\\Application\\chrome.exe")

4. Windows Defender Blocking Rod

Problem: "Operation did not complete successfully because the file contains a virus or potentially unwanted software"

Solution:

# Add exclusion to Windows Defender
Add-MpPreference -ExclusionPath "$env:TEMP\leakless-*" -Force

# Or manually:
# 1. Open Windows Security
# 2. Go to Virus & threat protection → Virus & threat protection settings
# 3. Click "Manage settings" under "Virus & threat protection settings"
# 4. Scroll down to "Exclusions"
# 5. Click "Add exclusions"
# 6. Select "Folder" and add: C:\Users\YourUsername\AppData\Local\Temp\leakless-*

5. "Login Failed: email input not found"

Problem: Email input field selector is not found

Solution:

# LinkedIn HTML may have changed
# Edit internal/automation/login.go and update the selector
emailInput, err := page.Element(`input#username`)

# To find the correct selector:
# 1. Open LinkedIn login page in Chrome
# 2. Right-click on email field → Inspect
# 3. Find the correct ID or class name
# 4. Update the selector in login.go

6. No Mouse Movements or Scrolling Visible

Problem: Application runs but mouse movements and scrolling not visible

Solution:

# Check 1: Browser must be visible (non-headless)
# In browser.go, ensure:
Headless(false)  # NOT true

# Check 2: Verify stealth actions are being called
# Check logs for "Executing random mouse movements..."
# If missing, check main.go has this line:
browser.PerformStealthActions(page)

# Check 3: Increase delay times in stealth files
# Edit internal/stealth/mouse.go and increase:
time.Sleep(time.Duration(1000+r.Intn(1000)) * time.Millisecond)

# Check 4: Verify page is fully loaded
# Wait longer before performing stealth actions
stealth.RandomDelay(3000, 5000)  // 3-5 second wait

7. "Login verification timeout"

Problem: Application takes too long to verify login

Solution:

# Check internet connection - LinkedIn may be slow
# Increase timeout in internal/automation/login.go:
timeout := time.After(60 * time.Second)  // Increase to 60 seconds

# Check if LinkedIn is accessible
# Try opening LinkedIn in regular Chrome browser

8. "Module not found" error

Problem: linkedin-automation/internal/... module not found

Solution:

# Ensure you're in the correct directory
pwd  # Should end with linkedin-automation

# Reinstall dependencies
go mod tidy
go mod download

# Verify go.mod exists
ls -la go.mod

9. "Port 9222 already in use"

Problem: Another instance of the application is running

Solution:

# On Windows (PowerShell):
Get-Process | Where-Object {$_.ProcessName -like "*chrome*"} | Stop-Process

# On Linux/Mac:
pkill chrome

# Or just restart your terminal/PowerShell

10. Application runs but immediately exits

Problem: No error message but program closes immediately

Solution:

# Check that .env file has valid credentials
cat .env

# Check that credentials are correct
# Try logging in manually to verify they work

# Check for silent errors:
go run main.go 2>&1 | tee output.log
# This saves all output to output.log for review

Windows Defender Blocking Rod

If you see: Operation did not complete successfully because the file contains a virus or potentially unwanted software

Add an exclusion to Windows Defender:

Add-MpPreference -ExclusionPath "$env:TEMP\leakless-*" -Force

Or manually:

1. Open Windows Security
2. Go to Virus & threat protection settings
3. Add exclusion for %TEMP%\leakless-*

Browser Not Opening

• Ensure Chrome/Chromium is installed or let Rod download it
• Check that Headless(false) is set in browser.go
• Verify no firewall is blocking the browser process

Login Fails

• Verify credentials in .env file are correct
• Check that LinkedIn is accessible from your network
• Ensure the LinkedIn login page HTML selectors haven't changed
• Check console logs for specific error messages

No Mouse Movements or Scrolling Visible

• Ensure Headless(false) is set (browser must be visible)
• Check that PerformStealthActions(page) is being called in main.go
• Verify logger output shows stealth action messages
• Increase pause durations in stealth files if movements are too fast to see

Dependencies

• github.com/go-rod/rod - Browser automation library
• github.com/joho/godotenv - .env file loading
• github.com/ysmood/leakless - Process management (included with rod)

See go.mod for complete dependencies.

Development

Running with Debug Output

Add logging to see detailed execution:

logger.Info("Your debug message")

Adding New Features

1. Create new modules in the internal/ directory
2. Follow the existing logging patterns
3. Use stealth functions to maintain natural behavior
4. Test thoroughly before deployment

Extending Stealth Techniques

Add new stealth methods in internal/stealth/:

// Example: New stealth function
func RandomHover(page *rod.Page) {
    // Implementation
}

Then call from browser.PerformStealthActions().

Important Notes

⚠️ LinkedIn's Terms of Service: Ensure your automation complies with LinkedIn's ToS. Unauthorized automation may result in account suspension.

✅ Best Practices:

• Use delays between actions
• Don't perform bulk operations rapidly
• Respect rate limits
• Log all activities for compliance
• Test with a dedicated account first

Contributing

Feel free to submit issues and enhancement requests!

License

MIT License - see LICENSE file for details (if added)

Author

Created for LinkedIn automation tasks.

Resources

• Rod Documentation
• Go Language Docs
• Chrome DevTools Protocol
• LinkedIn Developer

Changelog

v1.0.0 (2025-12-24)

• Initial release
• LinkedIn login automation
• Mouse movement simulation
• Page scrolling simulation
• Fingerprint masking
• Comprehensive logging