Summary Deep Eye is an advanced AI-driven vulnerability scanner and penetration testing tool for security professionals. It automates bug hunting, intelligent payload generation, and professional reporting by integrating multiple LLM providers with comprehensive security testing modules, enabling efficient and context-aware security assessments.

How It Works The tool employs a modular architecture, integrating with AI providers like OpenAI, Grok, OLLAMA, and Claude to power intelligent payload generation and context-sensitive analysis. It performs comprehensive scans using over 45 attack methods, covering web vulnerabilities, API security (OWASP API Top 10 2023), GraphQL, and business logic flaws. The workflow includes reconnaissance, targeted module execution, AI-assisted payload crafting, and report generation.

Quick Start & Requirements

• Installation: Recommended: Use platform-specific scripts (scripts/install.ps1 for Windows, scripts/install.sh for Linux/Mac). Manual: git clone, pip install -r requirements.txt, then configure config/config.yaml with AI provider API keys.
• Prerequisites: Python 3.8+, pip, and API keys for at least one supported AI provider (OpenAI, Anthropic/Claude, Grok, OLLAMA).
• Usage: Run via CLI: python deep_eye.py -u <target_url> or python deep_eye.py -c <config_file>. Scan parameters are primarily managed via config/config.yaml.
• Docs: Quick Start Guide, Architecture, and Testing procedures are available in the docs/ directory.

Highlighted Details

• Multi-AI Integration: Dynamic switching between OpenAI, Grok, OLLAMA, and Claude for enhanced AI capabilities.
• Comprehensive Modules: Supports 45+ attack vectors, including SQLi, XSS, SSRF, LFI/RFI, SSTI, API security, GraphQL, business logic flaws, and file upload vulnerabilities.
• AI-Powered Payloads: Generates CVE-aware, context-sensitive payloads with advanced obfuscation for WAF bypass.
• Advanced Reconnaissance & Reporting: Integrates OSINT, DNS enumeration, and produces professional PDF, HTML, or JSON reports.
• Extensibility: Features a custom plugin system and multi-channel notifications (Email, Slack, Discord).

Maintenance & Community Contributions are welcomed via Pull Requests, with support available through GitHub issues. The README does not specify active maintainers, sponsorships, or dedicated community channels like Discord or Slack.

Licensing & Compatibility Licensed under the MIT License. This permissive license allows for modification, distribution, and commercial use, generally compatible with closed-source projects.

Limitations & Caveats PDF report generation may face Windows-specific issues (though ReportLab is the default fallback). AI provider connection errors require verification of API keys and network access. Scanning performance can be impacted by target rate limiting or Web Application Firewalls (WAFs).