Skip to content

Architecture Overview

CyberCage provides comprehensive security through a simple, effective architecture that protects your development environment.

System Components

CyberCage Agent

A lightweight service running on your development machine that:

  • Automatically detects and protects MCP-enabled applications
  • Routes MCP communications through security checks
  • Syncs with cloud services for updates

CyberCage Cloud

The integrated cloud platform combines backend services and web dashboard to provide:

Security Services:

  • Advanced threat analysis and detection
  • Organization-wide server catalog management
  • Security policy enforcement
  • Threat intelligence updates
  • Audit logging and compliance

Dashboard Interface:

  • View and manage protected applications
  • Approve or block MCP servers
  • Review threat reports and security alerts
  • Manage team members and permissions
  • Configure integrations and notifications
  • Access analytics and metrics

How Protection Works

Step 1: Application Detection

When you install CyberCage, the agent automatically discovers MCP-enabled applications on your machine and begins protecting them immediately.

Step 2: Request Interception

Every MCP request is intercepted and analyzed before reaching the server:

  • Known safe operations proceed normally
  • Suspicious activities are blocked
  • Unknown servers require approval

Step 3: Response Validation

Server responses are validated before returning to your application:

  • Credentials and secrets are protected
  • Malicious content is filtered
  • Data integrity is maintained

Step 4: Continuous Monitoring

The system continuously:

  • Updates threat patterns
  • Syncs approved servers across your organization
  • Maintains audit trails

Security Model

Defense in Depth

Multiple layers of security work together:

  1. Server Approval - Only approved servers can run
  2. Request Analysis - Every request is evaluated for threats
  3. Response Filtering - Responses are sanitized for safety
  4. Policy Enforcement - Organization policies are applied

Deployment Flexibility

CyberCage adapts to your needs:

Standard Deployment

  • Full protection with all features
  • Real-time threat analysis
  • Complete audit trails
  • Ideal for most teams

Enterprise Deployment

  • Custom policy configuration
  • Advanced compliance features
  • SSO integration
  • Dedicated support

Integration Points

CyberCage seamlessly integrates with:

  • Development Tools - All major IDEs and editors
  • AI Assistants - Claude, GitHub Copilot, and others
  • Communication - Slack, Microsoft Teams
  • Monitoring - Webhooks for custom integrations
  • Compliance - Export capabilities for audit requirements

Performance Characteristics

CyberCage is designed to be invisible:

  • Minimal resource usage
  • Negligible latency impact
  • Intelligent caching
  • Background operations

Next Steps

Built in Berlin, DE 🇩🇪