X-CloudSentinel 🛡️

AI-Native Cloud Security Analysis through Graph Neural Networks and Contextual Transformers.

X-CloudSentinel is a research-grade security extension for VS Code that elevates Infrastructure-as-Code (IaC) auditing to the state-of-the-art. Unlike traditional static analysis tools, X-CloudSentinel uses Graph Neural Networks (GNN) to understand relational risks and BERT-based NER to detect secrets in context.

🚀 Key Features

🕸️ Relational IaC Scan (GNN)

Analyze Terraform and Kubernetes manifests as a property graph. Detect "transitive" vulnerabilities where a misconfiguration in one resource exposes another.

🔑 Contextual Secret Detection (BERT-NER)

Move beyond regex. Our Transformer-based NER model understands the programmatic context of your code, distinguishing between random GUIDs and actual high-risk credentials.

📊 Security Quality Index (SQI)

A novel, all-in-one security metric (0-100) that summarizes your project's risk posture across Secret Leakage, Misconfigurations, Access Risks, and Configuration Entropy.

🤖 Autonomous Agentic Reasoning

• Attack Propagation Agent: Detects how deep a vulnerability spreads.
• SQI Maximization Agent: Ranks fixes by their objective impact on security.
• Self-Reflection Agent: Monitors AI confidence to ensure high-trust results.

🛡️ Privacy-First & Offline

All AI models run locally on your machine. X-CloudSentinel never sends your code to external servers (Zero-Trust AI).

🛠️ Commands & Shortcuts

• Ctrl + Shift + A: [SOTA] Run Advanced Agentic Scan — Generates a comprehensive, explainable security report in the Output channel.
• X-CloudSentinel: Show Security Dashboard: Opens the rich, visual dashboard with SQI gauges and SHAP explanations.
• X-CloudSentinel: Scan Current File: Performs a standard real-time check.

📦 Requirements

• X-CloudSentinel Backend: Requires the local Python backend to be running for AI features.
  • Clone: https://github.com/X-CloudSentinel/X-CloudSentinel
  • Run: python app.py

🎓 Research Credits

X-CloudSentinel is a product of advanced research into GNNs, Transformers, and Bayesian Uncertainty in Cloud Security. It represents a "Peak SOTA" implementation for AI-Native DevSecOps.

Developed with ❤️ for the full-spectrum security researcher.