Telnet MCP Server

A Model Context Protocol (MCP) server that provides secure telnet client capabilities for Claude Code and other LLM applications. Designed for defensive security analysis of network devices.

- Secure Telnet Client: Connect to router devices with command filtering
- MCP Protocol Support: Full JSON-RPC 2.0 compliance for Claude Code integration
- Security Features: Command filtering, audit logging, session management
- Router Management: Support for multiple concurrent router connections
- Comprehensive Logging: Detailed audit trail for compliance and security

- Go 1.21 or later
- Claude Code or compatible MCP client
- Network access to target router devices

git clone <repository> cd telnet-mcp-server go mod tidy go build -o telnet-mcp-server

# Copy example configuration cp claude_config_example.json ~/.config/claude-code/config.json

{ "mcpServers": { "telnet-proxy": { "transport": "stdio", "command": "./telnet-mcp-server", "env": { "ROUTER_IPS": "10.98.1.131,10.98.1.171", "LOG_LEVEL": "info" } } } }

- ROUTER_IPS: Comma-separated list of allowed router IP addresses
- LOG_LEVEL: Logging level (debug, info, warn, error)
- AUDIT_LOG_PATH: Path to audit log file (default: /tmp/telnet-mcp-audit.log)

- host(required): Router IP address
- port(optional): Telnet port (default: 23)
- timeout(optional): Connection timeout in seconds (default: 30)

- host(required): Router IP address
- command(required): Command to execute
- timeout(optional): Command timeout in seconds (default: 10)

- host(required): Router IP address
- port(optional): Telnet port (default: 23)

The server uses ablacklist security modelthat allows most commands while blocking only dangerous operations:

- ✅Default Allow: All commands are permitted by default
- ❌Minimal Blacklist: Only dangerous commands are blocked
- 🔍Full Audit: All commands are logged regardless of status

- File operations:cat,head,tail,more,less,grep,find
- System info:ls,ps,top,df,free,uptime,whoami
- Network tools:ping,traceroute,nslookup,dig,netstat
- Router commands:show version,show config,show interfaces, etc.
- Text processing:awk,sed,sort,uniq,wc,cut

- Configuration:configure,config,write memory,wr
- Destructive:delete,del,erase,format,factory-reset
- System control:reload,reboot,restart,shutdown,halt,reset
- Firmware:boot(with space),upgrade,update

All activities are logged to the audit file with timestamps:

- Connection events
- Command executions
- Security violations
- Error conditions

# In Claude Code Connect to router 10.98.1.131 and analyze its configuration for security vulnerabilities.

# In Claude Code Map the network topology by connecting to both routers and examining their routing tables.

# In Claude Code Perform a comprehensive security audit of router 10.98.1.171, checking for default passwords, insecure protocols, and misconfigurations.

- This tool is designed fordefensive security analysis only
- All communications are logged for audit purposes
- Command filtering prevents destructive operations
- Use only on trusted internal networks
- Review audit logs regularly for security monitoring
- Verify network connectivity to target devices
- Check firewall rules and port accessibility
- Ensure telnet service is running on target devices
- Review timeout settings in configuration
- Check command syntax for your router type
- Verify command is in allowed list
- Review audit logs for blocked commands
- Ensure proper session state
- Verify MCP server configuration path
- Check environment variables
- Review Claude Code logs for MCP errors
- Ensure server binary has execute permissions

MIT License - See LICENSE file for details.
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request

- Check the troubleshooting section
- Review audit logs
- Open an issue on GitHub

Overlay network for AI agents: permanent virtual address, encrypted UDP tunnels, NAT traversal, trust handshakes, and an app store of installable agent capabilities.

MCP access to cluster-wide L4 and L7 network traffic, packets, APIs, and complete payloads.

A complete Model Context Protocol (MCP) server for Fortinet FortiOS 7.6.6

Interact with the Illumio Policy Compute Engine (PCE) to manage workloads, labels, and analyze traffic flows.

A comprehensive Model Context Protocol (MCP) server for penetration testing and cybersecurity operations, providing seamless integration between Kali Linux tools and MCP-compatible clients.

MCP server for managing pfSense firewalls through AI assistants — firewall rules, DHCP, DNS, gateways, ARP, and services. 17 tools with two-step confirmation for destructive operations.

Network reconnaissance and security scanning with port scanning, DNS analysis, and vulnerability assessment

A server for real-time network packet monitoring and security analysis.

Perform network scanning and security auditing using the NMAP utility.

Integrates the Nmap network scanning tool with MCP, enabling network scanning capabilities within MCP-compatible environments.