Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Secure MCP Server check the network interfaces on the Kali host"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A highly secure, enterprise-grade Model Context Protocol (MCP) server designed specifically for Kali Linux. This project bridges AI agents (via Termux/Android) to a secure local environment, enforcing strict authorization, modern cryptographic standards, and a zero-trust architecture.
Strict Least-Privilege Execution: Built-in safeguards against arbitrary code execution. Only explicitly allowlisted tools (e.g., specific network interface checks) are permitted. No generic shell execution (bash, exec, etc.) is allowed.
Robust Authentication (OAuth 2.0): Implements dynamic client registration, authorization code flow with PKCE (S256), and secure token exchange.
Modern Cryptography: Enforces ECDSA P-256 TLS for all communications with strict certificate pinning.
Audit & Logging: Comprehensive audit logs for all token exchanges, authorization requests, and tool executions. No credentials or bearer tokens are ever written to logs.
Network Isolation: Designed to run entirely on a local network (LAN) behind UFW rules, ensuring the server is not exposed to the public internet.
Related MCP server: Kali Factory MCP Server
The system operates on a localized Android-to-Kali Linux pipeline:
Android / Termux Client
│
│ LAN / Wi-Fi (TLS 1.3 Pinned)
▼
Kali Linux (Host: 0.0.0.0:8000)
│
▼
Secure MCP Server (Python 3.14.6)
│
├── Authentication & PKCE Validation
├── Authorization & Scope Checks
├── Audit Logging
└── Allowlisted Read-Only ToolsThis server cannot be deployed
Zero-trust gateway for AI agents: score tool calls, verify agent cards, enforce policy, audit.
OAuth 2.1 short-link tools for AI agents with scoped tokens, approvals, audit logs, and revocation.
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Security gateway for AI agents: policy, approval, and audited execution, no secrets shared.
