G3r4ki - AI-Powered Cybersecurity Operations Platform

Advanced AI-powered cybersecurity platform combining offensive and defensive security capabilities with intelligent, multi-provider AI integration.

Features • Installation • Quick Start • CLI Reference • Web Interface • Advanced Features

📋 Overview

G3r4ki is a comprehensive cybersecurity operations platform that combines:

AI-Powered Security Operations: Multi-provider AI integration (OpenAI GPT-4o, Anthropic Claude, DeepSeek) with local AI support
Interactive CLI: Command-line interface for security scanning, reconnaissance, and vulnerability assessment
Web Dashboard: Real-time visualization and control through Flask-based web interface
Autonomous Agents: Agent-based security operations for automated penetration testing and reconnaissance
Offensive Security Framework: Comprehensive modules for red team operations, exploit execution, and C2 infrastructure
Dual Operating Mode: Works with real security tools OR provides simulated responses for demonstration/training

Architecture

The platform operates in both online (cloud AI) and offline (local models) modes, making it suitable for diverse environments including air-gapped networks. It features:

Multi-AI Provider Proxy: Automatic provider selection and failover
Agent-Based Architecture: Self-planning autonomous security agents
Modular Offensive Framework: Mission-driven module selection (stealth, loud, persistence, data extraction)
Real-time Communication: WebSocket support for live updates
PostgreSQL Persistence: Database tracking for operations, agents, activities, and findings

✨ Features

Core Capabilities

🖥️ Interactive CLI

Network Scanning: Nmap-powered port scanning and service detection
Reconnaissance: WHOIS, DNS enumeration, subdomain discovery
Vulnerability Assessment: Nikto web scanning, SSL/TLS analysis
AI Query Interface: Natural language security assistance
Simulated Mode: Realistic mock data when tools aren't installed

🤖 AI Integration

Cloud Providers: OpenAI (GPT-4o), Anthropic (Claude 3.5 Sonnet), DeepSeek
Local Models: llama.cpp, vLLM (GPU-accelerated), GPT4All
Multi-Provider Queries: Simultaneous queries across all providers
Automatic Fallback: Smart provider selection based on availability

🌐 Web Dashboard

AI Assistant: Interactive chat with provider selection
Penetration Testing Tools: Reverse shell generator, exploit templates
Agent Management: Create, monitor, and control security agents
Mission Planning: Predefined mission profiles for operations
Analytics Dashboard: Real-time metrics and statistics
Logs Viewer: Comprehensive activity logging with filtering
Tool Management: Install and manage security tools

🎯 Offensive Security Modules

13 module categories with 100+ templates:

Credential Harvesting
Session Management
Post-Exploitation
Persistence Mechanisms
Evasion Techniques
Remote Execution
Data Exfiltration
RAT Deployment
Surveillance
Command & Control
Exploit Execution
Lateral Movement
Shell Generation

🛡️ Incident Response

Incident simulation framework
Response playbook execution
Performance evaluation
Automated report generation
Timeline reconstruction

🚀 Installation

Prerequisites

Required:

Python 3.8 or higher
PostgreSQL database (automatically available in Replit environment)
Linux or macOS operating system

Optional (for full functionality):

Security tools: nmap, nikto, sslscan, whois, dig
GPU with CUDA (for vLLM acceleration)
Internet access (for cloud AI providers)

Step 1: Clone and Setup

# Clone the repository
git clone <repository-url>
cd g3r4ki

# Install Python dependencies
pip install -r requirements.txt

Step 2: Environment Configuration

Create a .env file with your configuration:

cp .env.example .env

Edit .env and add your API keys:

OPENAI_API_KEY=sk-your-openai-key-here
ANTHROPIC_API_KEY=sk-ant-your-anthropic-key-here
DEEPSEEK_API_KEY=your-deepseek-key-here

DATABASE_URL=postgresql://user:password@localhost/g3r4ki

Step 3: Database Setup

The database will be automatically initialized on first run. If using PostgreSQL:

createdb g3r4ki

Step 4: (Optional) Install Security Tools

For real scanning (not simulated), install security tools:

Ubuntu/Debian:

sudo apt update
sudo apt install -y nmap nikto whois dnsutils

macOS:

brew install nmap nikto bind

Note: If tools aren't installed, G3r4ki automatically provides realistic simulated responses for demonstration purposes.

🎯 Quick Start

Interactive CLI Mode

Launch the interactive command-line interface:

python g3r4ki.py interactive

Example Session:

g3r4ki> scan example.com
Nmap scan of example.com (SIMULATED)
Generated: 2025-10-13 12:51:12
============================================================
Host: example.com

Open Ports:
  22/tcp: ssh (OpenSSH 8.9)
  80/tcp: http (nginx 1.18.0)
  443/tcp: https (nginx 1.18.0)

Note: This is simulated data. Install nmap for real scanning.

g3r4ki> query "What ports are commonly used for web services?"
Response:
------------------------------------------------------------
Web services commonly use:
- Port 80: HTTP (unencrypted web traffic)
- Port 443: HTTPS (encrypted web traffic)
- Port 8080: Alternative HTTP port
- Port 8443: Alternative HTTPS port
------------------------------------------------------------

Web Interface Mode

Launch the web dashboard:

python run_web_direct.py

Access the interface at: http://localhost:5000

The web interface provides:

Real-time AI chat interface
Visual penetration testing tools
Agent deployment and monitoring
Mission planning and execution
Analytics and reporting

📖 CLI Reference

Available Commands

`scan <target>`

Performs network port scanning using Nmap.

Usage:

g3r4ki> scan 192.168.1.1
g3r4ki> scan example.com

Simulated Output (when nmap not installed):

Host: example.com
Open Ports:
  22/tcp: ssh (OpenSSH 8.9)
  80/tcp: http (nginx 1.18.0)
  443/tcp: https (nginx 1.18.0)

Real Output (with nmap installed): Actual nmap scan results with service detection and version information.

`recon <target>`

Performs reconnaissance using WHOIS, DNS enumeration, and port scanning.

Usage:

g3r4ki> recon example.com

Simulated Output:

Reconnaissance Results for example.com
Generated: 2025-10-13 12:51:12
============================================================

WHOIS Information:
  Domain: example.com
  Registrar: Example Registrar Inc.
  Created: 1995-08-14
  Expires: 2025-08-13
  Name Servers: ns1.example.com, ns2.example.com

DNS Records:
  A: 93.184.216.34
  AAAA: 2606:2800:220:1:248:1893:25c8:1946
  MX: mail.example.com (priority: 10)
  NS: ns1.example.com, ns2.example.com

`vuln <target>`

Performs vulnerability scanning using Nikto and SSL/TLS analysis.

Usage:

g3r4ki> vuln https://example.com

Simulated Output:

Vulnerability Scan Results for https://example.com
Generated: 2025-10-13 12:51:12
============================================================

Web Vulnerabilities (Nikto):
  ✓ No missing security headers detected
  ! Server version disclosure: nginx/1.18.0
  ✓ HTTPS properly configured

SSL/TLS Analysis (SSLScan):
  Protocol Support:
    ✓ TLSv1.3: Enabled (secure)
    ✓ TLSv1.2: Enabled (secure)
    ✗ TLSv1.1: Disabled (good)
    ✗ TLSv1.0: Disabled (good)
  
  Certificate Information:
    Subject: CN=example.com
    Issuer: Let's Encrypt Authority X3
    Valid: 2024-01-15 to 2025-04-15
    ✓ Certificate is valid

`query <question>`

Query the AI assistant for security advice and analysis.

Usage:

g3r4ki> query "How do I detect SQL injection vulnerabilities?"
g3r4ki> query "Explain the OWASP Top 10"

Output: AI-generated response from your configured provider (OpenAI/Anthropic/DeepSeek).

`help`

Display available commands and usage information.

`exit` / `quit`

Exit the G3r4ki CLI.

🌐 Web Interface

Dashboard Tabs

1. AI Assistant

Multi-provider AI chat interface
Select between OpenAI, Anthropic, DeepSeek, or local models
System prompt customization
Query all providers simultaneously

2. Penetration Testing

Reverse Shell Generator: Generate shells for multiple languages (Python, Bash, PowerShell, PHP, Ruby, Perl)
Shell Variants: Basic, encoded, obfuscated options
Exploit Templates: Quick access to common exploit patterns
Payload Generation: Custom payload creation

3. Agent Operations

Create Agents: Deploy autonomous security agents
Agent Types: Pentest, reconnaissance, security assessment
Monitor Status: Real-time agent activity tracking
Control: Start, stop, configure agents

4. Mission Management

Mission Profiles:
- Stealth: Low-noise reconnaissance and enumeration
- Loud: Aggressive scanning and exploitation
- Persistence: Establish long-term access
- Data Extraction: Exfiltrate sensitive information
Planning: Define objectives and scope
Execution: Automated mission execution
Reporting: Detailed mission reports

5. Analytics

Agent activity metrics
Operations statistics
Tool installation status
Activity timeline
Performance indicators

6. Logs Viewer

Filter by operation, agent, type, status
Real-time log streaming
Search functionality
Auto-refresh options
Export capabilities

7. Tools Management

Categories: Recon, Remote Access, Defensive, Automation, Offensive, Data Analysis, Threat Intel, Red Team
Tool Status: Check installed tools
Installation: One-click tool installation
27+ Security Tools across 9 categories

🔧 Configuration

AI Provider Setup

OpenAI (GPT-4o)

Get API key from: https://platform.openai.com/api-keys
Add to .env: OPENAI_API_KEY=sk-...

Anthropic (Claude 3.5 Sonnet)

Get API key from: https://console.anthropic.com/
Add to .env: ANTHROPIC_API_KEY=sk-ant-...

DeepSeek

Get API key from: https://platform.deepseek.com/
Add to .env: DEEPSEEK_API_KEY=...

Local AI Models (Offline Operation)

llama.cpp:

# Install llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make

# Download a model
wget https://huggingface.co/model.gguf

# Configure in config.yaml
llm:
  local_models:
    llama_cpp:
      enabled: true
      model_path: /path/to/model.gguf

vLLM (GPU Required):

pip install vllm
# Configure GPU settings in config.yaml

Database Configuration

PostgreSQL connection string in .env:

DATABASE_URL=postgresql://username:password@localhost:5432/g3r4ki

In Replit environment, DATABASE_URL is automatically configured.

🎓 Operating Modes

Simulated Mode (Default)

When security tools are not installed, G3r4ki provides:

✅ Realistic simulated responses
✅ Educational examples
✅ Training scenarios
✅ Demonstration capabilities
✅ No external dependencies required

Use Cases:

Learning cybersecurity concepts
Training environments
Demonstrations
Development and testing

Real Mode (Production)

When security tools are installed, G3r4ki uses:

✅ Actual nmap network scanning
✅ Real DNS/WHOIS queries
✅ Live vulnerability assessment
✅ Production-ready results
✅ Comprehensive security analysis

Use Cases:

Professional penetration testing
Security assessments
Real-world operations
Production security audits

Note: All simulated outputs are clearly labeled to avoid confusion.

🚀 Advanced Features

Autonomous Security Agents

Create self-planning agents for automated operations:

# Via Web Interface: Agent Operations tab
# Or programmatically:
from src.agents.manager import AgentManager

manager = AgentManager()
agent = manager.create_agent(
    agent_type="pentest",
    name="Web App Tester",
    description="Automated web application security assessment"
)

agent.set_target("https://target.com")
agent.set_objective("Find OWASP Top 10 vulnerabilities")
result = manager.run_agent(agent.agent_id)

Offensive Security Modules

Access 100+ offensive modules across 13 categories:

Example - Credential Harvesting:

from src.offensive.module_loader import ModuleLoader

loader = ModuleLoader()
modules = loader.get_modules_by_category("credential_harvesting")
module = loader.load_module("mimikatz_dumper")
module.execute(target="192.168.1.100")

Incident Response Simulation

Simulate security incidents and test response:

from src.incident_response.simulator import IncidentSimulator

simulator = IncidentSimulator(config)
simulation = simulator.run_simulation(
    incident_type="ransomware",
    severity="critical"
)
report = simulator.generate_report(simulation)

Voice Command Interface (Experimental)

Control G3r4ki with voice commands:

python g3r4ki.py voice

Uses Whisper.cpp for speech-to-text and Piper for text-to-speech.

🛠️ Troubleshooting

Common Issues

1. CLI shows "tool not available" warnings

Solution: This is normal behavior. G3r4ki provides simulated responses when tools aren't installed. To use real tools:

sudo apt install nmap nikto whois dnsutils

2. AI providers not responding

Solution: Check your API keys in .env:

# Verify keys are set
cat .env | grep API_KEY

# Test connectivity
python -c "import openai; print('OpenAI OK')"

3. Database connection errors

Solution: Ensure PostgreSQL is running and DATABASE_URL is correct:

# Check PostgreSQL status
sudo systemctl status postgresql

# Test connection
psql $DATABASE_URL -c "SELECT 1;"

4. Web interface not loading

Solution: Check if port 5000 is available:

# Check port
lsof -i :5000

# Try alternative port
python run_web_direct.py --port 5001

5. Import errors or missing modules

Solution: Reinstall dependencies:

pip install -r requirements.txt --upgrade

FAQ

Q: Do I need all three AI providers?
A: No. You can use any combination. The platform works with one or more providers configured.

Q: Can I use G3r4ki offline?
A: Yes. Configure local AI models (llama.cpp, GPT4All, vLLM) for complete offline operation.

Q: Is simulated mode accurate?
A: Simulated responses are realistic examples for educational purposes but should not be used for actual security assessments.

Q: How do I add custom offensive modules?
A: Place your module in src/offensive/modules/<category>/ following the module template structure.

Q: Can I deploy agents to remote systems?
A: Yes. Configure SSH credentials in the agent configuration for remote deployment.

📝 Security & Legal

⚠️ Important Disclaimer

G3r4ki is designed for:

✅ Authorized security testing
✅ Educational purposes
✅ Research and development
✅ Security training

Unauthorized use is illegal and unethical.

Best Practices

Always obtain written authorization before testing any systems
Scope your testing to authorized targets only
Document all activities for compliance and reporting
Follow responsible disclosure for any vulnerabilities found
Comply with local laws and regulations (CFAA, Computer Misuse Act, etc.)

Data Privacy

G3r4ki stores operational data locally in PostgreSQL. When using cloud AI providers, prompts are sent to third-party services. Review provider privacy policies:

🤝 Contributing

Contributions are welcome! Areas for improvement:

Additional offensive modules
New AI provider integrations
Enhanced visualization features
Documentation improvements
Bug fixes and optimizations

📄 License

For educational and authorized security testing purposes only.

Unauthorized access to computer systems is illegal. The authors assume no liability for misuse of this software.

Name		Name	Last commit message	Last commit date
Latest commit History 15 Commits
.config/g3r4ki		.config/g3r4ki
src		src
.env.example		.env.example
.gitignore		.gitignore
README.md		README.md
g3r4ki.py		g3r4ki.py
pyproject.toml		pyproject.toml
requirements.txt		requirements.txt
run_web_direct.py		run_web_direct.py

RafalW3bCraft/WhisperAiEngine

Folders and files

Latest commit

History

Repository files navigation

G3r4ki - AI-Powered Cybersecurity Operations Platform

📋 Overview

Architecture

✨ Features

Core Capabilities

🖥️ Interactive CLI

🤖 AI Integration

🌐 Web Dashboard

🎯 Offensive Security Modules

🛡️ Incident Response

🚀 Installation

Prerequisites

Step 1: Clone and Setup

Step 2: Environment Configuration

Step 3: Database Setup

Step 4: (Optional) Install Security Tools

🎯 Quick Start

Interactive CLI Mode

Web Interface Mode

📖 CLI Reference

Available Commands

scan <target>

recon <target>

vuln <target>

query <question>

help

exit / quit

🌐 Web Interface

Dashboard Tabs

1. AI Assistant

2. Penetration Testing

3. Agent Operations

4. Mission Management

5. Analytics

6. Logs Viewer

7. Tools Management

🔧 Configuration

AI Provider Setup

OpenAI (GPT-4o)

Anthropic (Claude 3.5 Sonnet)

DeepSeek

Local AI Models (Offline Operation)

Database Configuration

🎓 Operating Modes

Simulated Mode (Default)

Real Mode (Production)

🚀 Advanced Features

Autonomous Security Agents

Offensive Security Modules

Incident Response Simulation

Voice Command Interface (Experimental)

🛠️ Troubleshooting

Common Issues

1. CLI shows "tool not available" warnings

2. AI providers not responding

3. Database connection errors

4. Web interface not loading

5. Import errors or missing modules

FAQ

📝 Security & Legal

⚠️ Important Disclaimer

Best Practices

Data Privacy

🤝 Contributing

📄 License

🔗 Resources

Learning Resources

Security Tools Documentation

AI Provider Documentation

About

Resources

Uh oh!

Stars

Watchers

Forks

`scan <target>`

`recon <target>`

`vuln <target>`

`query <question>`

`help`

`exit` / `quit`

Packages