Getting Started

Introduction

AgentShield provides comprehensive protection against AI agents and automated bots accessing your web applications. Whether you're protecting valuable content, preventing scraping, or ensuring fair usage of your services, AgentShield offers the tools you need.

How It Works

AgentShield uses multiple detection methods to identify automated traffic:

1. Behavioral Analysis - Monitors interaction patterns that differ from human users
2. Browser Fingerprinting - Identifies characteristics unique to automated browsers
3. Request Signatures - Analyzes HTTP headers and request patterns
4. Machine Learning - Continuously improves detection based on traffic patterns

Choose Your Implementation

AgentShield offers multiple integration options to fit your specific needs:

For Developers

JavaScript Beacon

Best for: Single-page applications, static sites, and traditional web apps

• Lightweight client-side detection
• WebWorker support for non-blocking execution
• Real-time event tracking
• Works with any JavaScript framework

Server-side Middleware

Best for: Full-stack applications with server control

• Next.js - Edge-optimized middleware with App Router support
• Express - Drop-in middleware for Node.js applications
• Complete request analysis before response
• Session tracking and advanced analytics

For Marketers

Marketing Pixel

Best for: Marketing teams using Google Tag Manager

• No coding required
• GTM template for easy installation
• Automatic pageview tracking
• Custom event configuration through GTM interface

Core Concepts

Detection Confidence

AgentShield provides a confidence score (0-1) for each detection:

• 0.0-0.3: Likely human
• 0.3-0.7: Uncertain (may be automated tools or assistive technology)
• 0.7-1.0: Likely AI agent or bot

Response Actions

You can configure how to handle detected agents:

• Allow - Log detection but allow access
• Block - Return 403 Forbidden response
• Redirect - Send to a different page
• Challenge - Request additional verification
• Custom - Define your own response logic

Privacy & Compliance

AgentShield is designed with privacy in mind:

• GDPR compliant
• Respects Do Not Track headers
• Configurable data retention
• No PII collection by default
• IP anonymization options

Installation Overview

The installation process varies by integration type:

1. Choose your package - Select the appropriate npm package
2. Install dependencies - Add to your project
3. Configure - Set up with your API key and preferences
4. Deploy - Push to production
5. Monitor - Track detections in your dashboard

System Requirements

Client-side (Beacon/Pixel)

• Modern browsers (Chrome 60+, Firefox 55+, Safari 11+, Edge 79+)
• JavaScript enabled
• Optional: WebWorker support for optimal performance

Server-side (Middleware)

• Node.js 16+
• Next.js 13+ (for Next.js integration)
• Express 4+ (for Express integration)

Quick Decision Guide

Not sure which integration to choose? Here's a quick guide:

Next Steps

Ready to get started? Follow our Quick Start Guide to have AgentShield running in under 5 minutes.

For a deeper understanding of how AgentShield works, check out our Core Concepts guide.