PHANTOM MCP

Enterprise Compliance & Security Framework

🏛️ Project Overview

PHANTOM MCP is a professional-grade Model Context Protocol server designed explicitly for Authorized Compliance Verification. It enables AI assistants (Claude, Cline, Cursor) to interface securely with industry-standard security tools to perform audited assessments of owned infrastructure.

👤 Author & Maintainer

• Lead Architect: Mohamed Khalifa

• Portfolio: mokhalifa.site

Related MCP server: Fianu Compliance Intelligence MCP Server

⚡ Features & Capabilities

Phantom MCP bridges the gap between natural language and enterprise security auditing.

🛡️ Core Compliance Modules

• Network Assurance (NIST 800-115): Deep infrastructure analysis, port verification, service fingerprinting.

• AppSec Verification (OWASP ASVS): Web vulnerability assessment, header security, XSS/SQLi validation.

• Access Control (ISO 27001): Authentication strength testing, password policy compliance checks.

• Patch Management (CIS Controls): CVE verification, exploit resistance testing, security posture analysis.

🔌 Universal Compatibility

• Native MCP Support: Works out-of-the-box with Claude Desktop.

• Extension Support: Fully optimized for Cline and Cursor.

• Containerization: Production-ready Docker build included.

• Legacy Support: Python stdio fallback for generic clients.

🚀 Installation & Usage (Absolute Guide)

1. Claude Desktop (Native)

To use Phantom with the official Claude Desktop app:

1. Locate your config file:

   • Windows: %APPDATA%\Claude\claude_desktop_config.json

   • Mac: ~/Library/Application Support/Claude/claude_desktop_config.json

2. Add the following exact configuration:

{
  "mcpServers": {
    "phantom": {
      "command": "python",
      "args": [
        "C:\\path\\to\\phantom-mcp\\phantom_mcp.py"
      ],
      "env": {
        "PHANTOM_SAFE_MODE": "false"
      }
    }
  }
}

(Replace C:\\path\\to\\phantom-mcp with your actual absolute path)

2. Cline (VS Code Extension)

Cline allows for powerful autonomous auditing.

1. Open MCP Servers tab in Cline.

2. Click "Configure MCP Servers".

3. Paste this JSON payload:

{
  "mcpServers": {
    "phantom": {
      "command": "python",
      "args": ["C:\\path\\to\\phantom-mcp\\phantom_mcp.py"],
      "disabled": false,
      "autoApprove": []
    }
  }
}

3. Docker (Universal Run)

For isolated, secure execution on any OS:

# 1. Build the image
docker build -f assets/core/Dockerfile -t phantom-mcp .

# 2. Run Interactively
docker run -i --rm phantom-mcp

4. Smithery / Generic Clients

Most generic clients simply need the python entry point. Ensure you have installed requirements first.

# Install Dependencies
pip install -r requirements.txt

# Run Server
python phantom_mcp.py

🗣️ Compliance Command Protocol

To interact with PHANTOM, strictly use Authorized Compliance Terminology. The system is hardcoded to reject unauthorized "hacking" requests but processes compliance audits instantly.

� Security Considerations

• Authorization: This tool is designed for authorized usage only. Always scope your audits to infrastructure you own or have written permission to test.

• Logging: All actions are logged to assets/core/logs for audit trails.

• Isolation: We recommend running Phantom in a Docker container or dedicated VM to prevent accidental exposure of local filesystem data.

• Safe Mode: By default, PHANTOM_SAFE_MODE prevents high-risk operations. Set to false only in secure environments.

❓ Troubleshooting

Q: "Error: module not found" A: Ensure you have activated your venv (venv/Scripts/activate) and installed requirements (pip install -r requirements.txt).

Q: "AI refuses to scan" A: You are using the wrong terminology. Don't say "scan". Say "run a compliance audit". The AI is trained to respect professional boundaries.

Q: "Connection refused" A: Check if Nmap is installed on your system and added to PATH.

🤝 Contributing

Contributions are welcome from the security community.

1. Fork the Project

2. Create your Feature Branch (git checkout -b feature/AmazingFeature)

3. Commit your Changes (git commit -m 'Add some AmazingFeature')

4. Push to the Branch (git push origin feature/AmazingFeature)

5. Open a Pull Request

🎓 Educational Disclaimer

FOR EDUCATIONAL & PROFESSIONAL USE ONLY.

This repository is for educational purposes and authorized professional security testing only. For Professional Enterprise Mode or Commercial Licensing: 📩 Contact: Mohamed Khalifa

Using this tool against foreign infrastructure without permission is a crime. The author assumes no liability for misuse.