Kali MCP Server

中文文档 | English

A Model Context Protocol (MCP) server that exposes Kali Linux security tools to Claude AI for authorized security testing and penetration testing.

• AI-Powered Assistant: Use plain English or Chinese to interact with Kali tools
• Just say: "scan 192.168.1.1" or "生成webshell 密码123"

• Reconnaissance: Network scanning, DNS enumeration, WHOIS lookups, HTTP headers
• Vulnerability Scanning: Nmap, Nikto, SSL/TLS scanning, directory brute forcing
• Exploitation: Metasploit framework, Hydra, sqlmap, weevely webshells
• Post-Exploitation: Webshell generation/management, privilege escalation, persistence
• Utilities: Base64, Hex, URL encoding, hashing, ROT13, XOR

• Built-in authorization checks for dangerous operations
• Comprehensive audit logging
• Network range restrictions (allow/block lists)
• Rate limiting to prevent abuse

• Kali Linux (or compatible Debian-based system)
• Python 3.11+
• SSH access (for remote deployment)
• Common Kali tools: nmap, metasploit-framework, hydra, nikto, weevely, sqlmap

# Clone the repository
git clone https://github.com/yourusername/kali-mcp.git
cd kali-mcp

# Run the installation script
chmod +x install_on_kali.sh
./install_on_kali.sh

# 1. Create virtual environment
python3 -m venv venv
source venv/bin/activate

# 2. Install dependencies
pip install -r requirements.txt

# 3. Install additional Python packages for weevely
pip install mako prettytable pyyaml jinja2 requests paramiko cryptography pillow

# 4. Start the server
python -m kali_mcp.server

authorization:
  allowed_ranges: []          # Leave empty to allow all, or specify networks
  blocked_ranges: []          # Networks to block (e.g., ["127.0.0.0/8"])
  require_auth_for_exploitation: true

security:
  rate_limit:
    enabled: true
    max_requests_per_minute: 60

logging:
  file: /var/log/kali_mcp.log
  level: INFO

Add to your MCP configuration:

For Local Connection:

{
  "mcpServers": {
    "kali": {
      "command": "python",
      "args": ["/path/to/kali-mcp/kali_mcp/server.py"],
      "env": {
        "LOG_LEVEL": "INFO"
      }
    }
  }
}

For Remote SSH Connection:

{
  "mcpServers": {
    "kali": {
      "command": "ssh",
      "args": [
        "user@your-kali-ip",
        "cd /path/to/kali-mcp && source venv/bin/activate && python -m kali_mcp.server"
      ],
      "env": {
        "LOG_LEVEL": "INFO"
      }
    }
  }
}

Just describe what you want in plain English or Chinese:

English Examples:

"Scan 192.168.1.1 for open ports"
"Generate a webshell with password: mypass123"
"Whois lookup for example.com"
"Search for vsftpd exploits"
"Brute force directories on http://example.com"
"Test SQL injection on http://example.com/page?id=1"

中文 Examples:

"扫描 192.168.1.1 的开放端口"
"生成webshell 密码123"
"查询 example.com 的whois信息"
"搜索 vsftpd 漏洞"
"爆破目录 http://example.com"

Or call specific tools directly:

# Port scanning
nmap_scan(target="192.168.1.1", ports="1-1000", scan_type="-sS")

# Webshell generation
generate_webshell(password="mypass123", filename="shell.php")

# WHOIS lookup
whois_lookup(domain="example.com")

# Exploit search
searchsploit(query="apache 2.4.49")

# SQL injection testing
sqlmap_scan(url="http://example.com/page?id=1")

• nmap_scan - Port and service scanning with Nmap
• ping_sweep - Discover live hosts on network
• whois_lookup - Domain/IP WHOIS information
• dns_enumerate - DNS record enumeration (A, AAAA, MX, NS, TXT, etc.)
• dns_brute_force - Subdomain brute forcing
• http_headers - Get HTTP headers from URL
• ssl_cert_info - SSL/TLS certificate information

• nikto_scan - Web server vulnerability scanner
• ssl_scan - SSL/TLS configuration scanning
• dir_brute_force - Directory and file brute forcing
• smb_enum - SMB shares enumeration

• searchsploit - Search Exploit-DB
• hydra_crack - Password brute forcing with Hydra
• john_crack - Password cracking with John the Ripper
• generate_payload - Generate payloads with msfvenom
• sqlmap_scan - SQL injection testing with sqlmap
• msf_exploit - Execute Metasploit exploits
• msf_auxiliary - Run Metasploit auxiliary modules
• wpscan_scan - WordPress vulnerability scanner
• enum4linux_scan - SMB/Windows enumeration

• generate_webshell - Generate webshells with weevely
• webshell_execute - Execute commands through webshell
• webshell_upload - Upload webshell to target
• suggest_privilege_escalation - Suggest privesc vectors
• suggest_persistence - Suggest persistence mechanisms

• base64_encode/decode - Base64 encoding/decoding
• url_encode/decode - URL encoding/decoding
• hex_encode/decode - Hexadecimal encoding/decoding
• hash_data - Hash data (MD5, SHA1, SHA256, SHA512)
• identify_hash - Identify hash type
• rot13 - ROT13 cipher
• xor_data - XOR encoding/decoding
• generate_random_string - Generate random strings
• system_health_check - Check system health and tool availability

• Dangerous operations (exploitation, post-exploitation) require explicit authorization
• Configurable danger levels for different operation types
• User confirmation prompts before critical actions

All operations logged with:

• Timestamp
• Operation type
• Target
• Parameters
• Results
• User identification

• Allowed ranges whitelist (optional)
• Blocked ranges blacklist
• CIDR notation supported
• Hostname bypass (when IP cannot be determined)

• Configurable requests per minute
• Prevents automated abuse
• Protects both your systems and targets

IMPORTANT: This tool is designed for authorized security testing only.

• ✅ Written permission from system owners
• ✅ Only test systems you own or are authorized to test
• ✅ Follow responsible disclosure practices
• ✅ Comply with all applicable laws and regulations
• ✅ Use appropriate rate limiting

• ❌ Unauthorized scanning of networks you don't own
• ❌ Testing systems without explicit permission
• ❌ Using for malicious purposes
• ❌ Violating any local or international laws

The developers are not responsible for misuse of this tool.

kali-mcp/
├── kali_mcp/
│   ├── __init__.py
│   ├── __main__.py
│   ├── server.py              # Main MCP server
│   ├── assistant.py           # Natural language AI assistant
│   ├── security.py            # Security & authorization
│   └── tools/
│       ├── __init__.py
│       ├── recon.py           # Reconnaissance tools
│       ├── scan.py            # Scanning tools
│       ├── exploit.py         # Exploitation tools
│       ├── post.py            # Post-exploitation tools
│       └── utils.py           # Utility functions
├── config.yaml                # Configuration file
├── requirements.txt           # Python dependencies
├── install_on_kali.sh         # Installation script
├── setup.py                   # Package setup
├── claude_config_example.json # Example MCP config
└── README.md                  # This file

Contributions are welcome! Please ensure:

1. Authorization Checks - All dangerous tools require authorization
2. Documentation - Comprehensive docstrings and comments
3. Error Handling - Graceful failure with helpful messages
4. Audit Logging - Log all operations
5. Testing - Test in safe environments first

1. Create function in appropriate tools/ module
2. Add security validation if needed
3. Register in server.py tool list
4. Add handler in handle_call_tool()
5. Update documentation

Tool not found:

sudo apt install nmap metasploit-framework hydra nikto weevely sqlmap

Permission denied:

• Run with appropriate privileges
• Configure sudoers for specific commands

SSH connection fails:

• Setup SSH key authentication
• Test: ssh user@kali-ip

weevely not working:

pip install mako prettytable pyyaml jinja2 requests

Enable debug logging:

LOG_LEVEL=DEBUG python -m kali_mcp.server

This project is provided as-is for authorized security testing purposes.

Use responsibly and legally.

• MCP (Model Context Protocol) by Anthropic
• Kali Linux team
• All open-source security tools integrated

For issues, questions, or contributions:

• Open an issue on GitHub
• Check existing documentation
• Review tool-specific help text

Remember: With great power comes great responsibility. Use this tool ethically and legally!