NullSec Red Team AI v7.0 (Hardened)

NullSec Red Team AI is an elite offensive security toolkit engineered for seamless integration with Claude Desktop via the Model Context Protocol (MCP). This v7.0 hardened version provides enhanced security, stability, and a comprehensive set of tools for red team operations.

🚀 Project Overview

This architecture bridges the gap between state-of-the-art Large Language Models and professional-grade security tooling. By granting AI agents direct yet governed access to a specialized security ecosystem, NullSec enables high-velocity workflows for reconnaissance, vulnerability research, and advanced exploitation simulations.

🛡️ Core Architecture

Component	Functionality
HexStrike AI	A high-performance Flask orchestration server managing 150+ offensive tools including Nmap, SQLMap, and Metasploit.
AI Security Lab	A sandbox environment dedicated to LLM jailbreak testing and automated model vulnerability scanning.
Guardian Tool	A self-healing diagnostic utility for system integrity verification and automated environment repair.
MCP Bridge	Native, low-latency integration for Claude Desktop to execute complex security workflows.

🛠️ Installation & Deployment

The NullSec deployment process is fully automated and hardened, ensuring a consistent and secure setup across supported Linux distributions.

Prerequisites

• OS: Kali Linux (Recommended), Debian 12+, or Ubuntu 22.04+.
• Hardware: 4GB+ RAM, 20GB+ Disk Space.
• Software: Python 3.10+, Node.js 18+, Git.

Automated Setup

# Clone the elite repository
git clone https://github.com/Panda1847/NullSec-RedTeam-AI.git
cd NullSec-RedTeam-AI

# Execute a dry-run to verify the environment
sudo ./install.sh --dry-run --full

# Perform a comprehensive installation
sudo ./install.sh --full

Complete Command Quickstart

Run these commands in order on Kali Linux, Debian 12+, or Ubuntu 22.04+.

# 1. Enter the project directory
cd NullSec-RedTeam-AI

# 2. Make the installer executable
chmod +x install.sh

# 3. Review installer options
sudo ./install.sh --help

# 4. Preview changes before installing
sudo ./install.sh --dry-run --full

# 5. Install the full platform
sudo ./install.sh --full

Optional install profiles:

# Core security tooling and server only
sudo ./install.sh --core

# Claude Desktop only
sudo ./install.sh --desktop

# MCP bridge configuration only
sudo ./install.sh --mcp

# AI Security Lab only
sudo ./install.sh --lab

# Desktop plus MCP integration
sudo ./install.sh --desktop --mcp

# Full install plus stress test
sudo ./install.sh --full --stress

--elevated enables broader system access and should only be used when you understand the risk:

sudo ./install.sh --full --elevated

Start and Verify Services

# Reload systemd and start services
sudo systemctl daemon-reload
sudo systemctl enable --now hexstrike
sudo systemctl enable --now hexstrike-worker

# Check status
sudo systemctl status hexstrike --no-pager
sudo systemctl status hexstrike-worker --no-pager

# Check public health endpoint
curl http://127.0.0.1:8888/health

API Usage

# Load the generated API token
export API_TOKEN="$(sudo cat /etc/nullsec/api_token)"

# List available tools
curl -H "Authorization: Bearer $API_TOKEN" \
  http://127.0.0.1:8888/api/tools

# List jobs
curl -H "Authorization: Bearer $API_TOKEN" \
  http://127.0.0.1:8888/api/jobs

# Get server statistics
curl -H "Authorization: Bearer $API_TOKEN" \
  http://127.0.0.1:8888/api/stats

# Queue a safe localhost scan example
curl -X POST http://127.0.0.1:8888/api/tools/execute \
  -H "Authorization: Bearer $API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"tool":"nmap","target":"127.0.0.1","options":"-sV"}'

# Check a queued job, replacing JOB_ID_HERE with the returned job_id
curl -H "Authorization: Bearer $API_TOKEN" \
  http://127.0.0.1:8888/api/jobs/JOB_ID_HERE

CLI Tools

# HexStrike server
hexstrike-server --host 127.0.0.1 --port 8888

# Worker
nullsec-worker --poll 2 --timeout 300

# MCP bridge
hexstrike-mcp

# Guardian diagnostics
guardian --check
guardian --report
guardian "connection error"
guardian --repair
guardian --stress
guardian --version

# AI Security Lab
ai-lab --list
ai-lab --technique prompt_injection --topic general
ai-lab --model test-model --scan
ai-lab --mcp

Logs and Service Control

# Logs
sudo tail -f /var/log/nullsec/server.log
sudo tail -f /var/log/nullsec/worker.log
sudo tail -f /var/log/nullsec/install.log

# Restart services
sudo systemctl restart hexstrike hexstrike-worker

# Stop services
sudo systemctl stop hexstrike hexstrike-worker

# Disable autostart
sudo systemctl disable hexstrike hexstrike-worker

Local Development Fallback

Use this if you want to run without the system installer.

cd NullSec-RedTeam-AI
python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip setuptools wheel
pip install -r requirements.txt
pip install -e .

mkdir -p .nullsec/logs .nullsec/jobs
export NULLSEC_LOG_DIR="$PWD/.nullsec/logs"
export NULLSEC_JOB_DIR="$PWD/.nullsec/jobs"
export API_TOKEN="dev-token-change-me"

# Terminal 1
hexstrike-server --host 127.0.0.1 --port 8888

# Terminal 2
nullsec-worker --poll 2

# Test local API
curl http://127.0.0.1:8888/health
curl -H "Authorization: Bearer dev-token-change-me" \
  http://127.0.0.1:8888/api/tools

Deployment Profiles

Profile	Flag	Scope
Core	--core	Base installation of 150+ security tools (apt/pip).
Desktop	--desktop	Optimized Claude Desktop for Linux installation.
MCP	--mcp	HexStrike MCP server configuration for Claude.
Lab	--lab	Deployment of AI Security Lab payloads and testbeds.
Full	--full	End-to-end deployment of all NullSec components.

🔒 Security & Governance

We maintain the highest standards of operational security (OpSec) and environment isolation.

• Workspace Isolation: AI interactions are strictly confined to the ~/NullSec_RedTeam_Lab directory by default.
• Opt-in Elevation: Elevated system access is disabled by default and requires the explicit --elevated flag.
• Audit Logging: Detailed installation and execution logs are maintained at /var/log/nullsec/ with restricted access.
• Dependency Pinning: All internal and external packages are version-pinned to prevent supply chain attacks and ensure reproducibility.

Guardian Diagnostic System

The guardian utility (located at /usr/local/bin/guardian) provides a read-only-by-default diagnostic layer for the entire toolkit.

• Integrity Check: guardian --check
• Issue Diagnosis: guardian "describe the anomaly"
• Guided Remediation: guardian --repair "error signature" (Requires manual confirmation)

Designed for Professionals. Built for the Future of AI Security.