SentinelAI is an advanced, all-in-one AI-powered offensive security framework, automated vulnerability scanner, and penetration testing arsenal. Combining instant reconnaissance, deep web vulnerability audits, automated threat modeling, CVSS 3.1 calculations, copy-paste developer remediation patches, an interactive AI Security Copilot, and a curated multi-tool launchpad into a seamless terminal and web experience.
███████╗███████╗███╗ ██╗████████╗██╗███╗ ██╗███████╗██╗ █████╗ ██╗
██╔════╝██╔════╝████╗ ██║╚══██╔══╝██║████╗ ██║██╔════╝██║ ██╔══██╗██║
███████╗█████╗ ██╔██╗ ██║ ██║ ██║██╔██╗ ██║█████╗ ██║ ███████║██║
╚════██║██╔══╝ ██║╚██╗██║ ██║ ██║██║╚██╗██║██╔══╝ ██║ ██╔══██║██║
███████║███████╗██║ ╚████║ ██║ ██║██║ ╚████║███████╗███████╗██║ ██║██║
╚══════╝╚══════╝╚═╝ ╚═══╝ ╚═╝ ╚═╝╚═╝ ╚═══╝╚══════╝╚══════╝╚═╝ ╚═╝╚═╝
⚡ AI-POWERED OFFENSIVE SECURITY & PENETRATION TESTING ARSENAL ⚡
SentinelAI is fully engineered to run seamlessly across Windows 10/11, Linux, and macOS:
- Windows Package Manager (
winget) & Chocolatey (choco): 1-click install industry-standard external tools directly from the terminal console. - Smart Path Resolver: Automatically locates tools in
C:\Program Files,C:\Program Files (x86), WinGet Links, Chocolatey bin, Scoop shims, and virtual environmentScripts/. - Zero-Setup SQLite Mode: Embedded database fallback ensures scans and API servers run natively on Windows without requiring Docker, PostgreSQL, or WSL.
- Windows Console ANSI & UTF-8: Automatic Virtual Terminal Processing (VT100) initialization guarantees crystal-clear cyberpunk colors, bold borders, and emoji icons in Windows Command Prompt (CMD), PowerShell, and Windows Terminal.
- Windows Celery Compatibility: Automatically activates the
solopool on Windows to eliminate Linuxfork()compatibility errors.
- Multi-Provider LLM Integration: Connects to Groq (
llama-3.3-70b-versatile), OpenAI (gpt-4o-mini), Venice AI (llama-3.3-70b), and DeepSeek (deepseek-chat) with automatic multi-provider failover. - Offline Heuristic Engine: Built-in comprehensive rule and pattern database guarantees full scoring and remediation capabilities with zero API keys required.
- Automated Holistic Threat Modeling: Correlates perimeter reconnaissance, open ports, technologies, and vulnerabilities into an executive posture rating (
CRITICAL,HIGH,MEDIUM), primary threat vectors, immediate 24-hour actions, and strategic roadmaps. - CVSS 3.1 Scoring & Vectors: Calculates standardized CVSS 3.1 vector strings (e.g.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) and exploit complexity metrics. - Developer Remediation Code Patches: Generates precise, copy-paste defensive patches for Python, PHP, JavaScript, SQL, and Nginx/Apache web server configurations.
- Context-Aware Technical Advisory: The Copilot automatically ingests the active target, detected WAF, open ports, and discovered vulnerabilities.
- On-Demand Penetration Testing Guidance: Ask the Copilot technical questions right inside the terminal:
- "How do I verify if this SQL injection is vulnerable to time-based exfiltration?"
- "Generate the secure Nginx configuration to enforce HSTS and CSP."
- "Explain the business impact of this CORS misconfiguration to an executive."
- Post-Scan 1-Click Consultation: Launch Copilot directly from the scan summary to immediately interrogate findings.
Instant host environment detection, 1-click auto-installation, live execution streaming, and evidence logging for 26+ industry-standard security tools across 10 categories:
- Network Discovery & Port Auditing: Nmap, RustScan, Masscan, Netcat (nc)
- Web Vulnerability & DAST Scanners: Nikto, Nuclei, Wapiti
- CMS & Framework Auditing: WPScan, CMSmap, Droopescan
- Directory, File & Parameter Fuzzing: Gobuster, FFuF, Dirsearch, Feroxbuster
- Database & SQL Injection Auditing: SQLmap, Ghauri
- OSINT & Subdomain Intelligence: Sublist3r, WhatWeb, Whois, theHarvester
- DNS & Infrastructure Enumeration: Dnsrecon, Dnsenum
- SSL / TLS & Cipher Auditing: testssl.sh, SSLyze
- Secrets & Repository Intelligence: TruffleHog
- Web Parameter & Route Spidering: ParamSpider
- Perimeter & Infrastructure Recon: DNS + crt.sh Subdomains + 22 Ports + WAF + Security Headers (~10s).
- Content & Sensitive File Discovery: Web Crawl + 50 Paths + Configs + Backups + LFI (~25s).
- API & Endpoint Attack Surface: Swagger/OpenAPI + GraphQL + CORS + SSRF Vectors (~20s).
- Complete Full-Scope Assessment: End-to-end full audit suite with AI correlation and false-positive elimination.
- Subdomain Discovery: High-speed discovery via Certificate Transparency (crt.sh) & asynchronous DNS resolution.
- Port Scanner: Multi-threaded TCP connect scanner targeting top operational service ports.
- WAF & Firewall Fingerprinting: Automatic detection for Cloudflare, AWS WAF, Akamai, Imperva, Sucuri, and ModSecurity.
- Tech Stack Identification: Identifies Nginx, Apache, WordPress, React, Next.js, Django, Laravel, and exposed server versions.
- SQL Injection (SQLi): Error-based, boolean-based, and union-based injection audit.
- Cross-Site Scripting (XSS): Reflected, DOM-based, and attribute-context script injection checks.
- Security Misconfigurations: Missing HSTS, CSP, X-Frame-Options clickjacking protection, and CORS reflection.
- Sensitive File & Information Disclosure: Audits
/.env,/.git, backup dumps, Swagger docs, and configuration files. - API & Authentication Auditing: Discovers REST/GraphQL endpoints, JWT signing flaws, and IDOR patterns.
- Cyberpunk Dark HTML Reports: Executive summary, AI threat model card, risk distributions, CVSS vector badges, and syntax-highlighted code patches.
- Markdown Reports: Formatted with executive summaries and developer patches, ready for HackerOne, Bugcrowd, or internal wikis.
- JSON Reports: Machine-readable data for automated DevSecOps pipelines.
Simply double-click install_windows.bat (or run in Command Prompt):
git clone https://github.com/Mazonia/sentinelai.git
cd sentinelai
install_windows.batThis automatically sets up Python virtual environments, installs requirements, registers global sentinelai commands, and launches the console.
git clone https://github.com/Mazonia/sentinelai.git
cd sentinelai
.\install_windows.ps1Once installed, simply type sentinelai or .\sentinelai.ps1 in any terminal!
git clone https://github.com/Mazonia/sentinelai.git
cd sentinelai
chmod +x install_linux.sh sentinel.sh
./install_linux.shThis automatically sets up the Python virtual environment, installs dependencies, registers global sentinelai commands in ~/.local/bin, and launches the framework!
git clone https://github.com/Mazonia/sentinelai.git
cd sentinelai
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -e .
sentinelaiHeadless Server Support: On headless Linux VPS/cloud servers (without a desktop GUI/browser), SentinelAI automatically detects the environment and provides direct absolute file paths for HTML/Markdown reports without throwing
xdg-openbrowser launch errors.
Simply type sentinelai (or sentinel) — no python prefixes or complex syntax needed:
sentinelaiThis launches the guided numbered console:
[1] ⚡ Multi-Task Assessment Workflows (1-Click Chained Presets: Perimeter, Content, API, Full)
[2] 🎯 Full Deep Vulnerability Scan (Crawl, OWASP Audit Suite, AI Threat Model & Report)
[3] 🔍 Reconnaissance & OSINT Engine (DNS, Subdomains, 22-Port Scanner, WAF Identification)
[4] 📂 Fast Sensitive File Fuzzer (Probes .env, .git, backups, SQL dumps, APIs)
[5] 🧪 Targeted Vulnerability Tester (Directly audit SQLi, XSS, CORS, SSRF, LFI, Headers)
[6] 🧰 Security Tools Arsenal (26+ Tools across 10 Categories with 1-Click Auto-Install & Logs)
[7] 📄 View & Open Audit Reports (Browse, read, and 1-click open Cyberpunk HTML reports)
[8] 🤖 Interactive AI Security Copilot (Terminal chat, threat reasoning, and remediation code)
[9] 🌐 Start SentinelAI REST API Server (Launch FastAPI backend on localhost:8000)
[10] 🎯 Set / Change Active Target (Configure session memory target)
[0] 🚪 Exit
Just enter a number (1, 2, 3...). SentinelAI guides you through each step and remembers your active target so you rarely have to re-type anything!
You can also run direct actions by passing arguments to sentinelai:
# Full Automated Scan with AI Remediation & Report Generation:
sentinelai scan https://example.com
# Launch Interactive AI Security Copilot:
sentinelai copilot
# Domain Reconnaissance & OSINT:
sentinelai recon example.com
# Sensitive File & Hidden Path Fuzzing:
sentinelai fuzz https://example.com
# Launch Security Tool Arsenal:
sentinelai tools
# Browse & open generated reports:
sentinelai reports
# Start REST API backend:
sentinelai apiTo enable live LLM reasoning, export any of the supported API keys:
# Groq (Ultra-fast Llama-3.3-70b - Recommended)
export GROQ_API_KEY="gsk_..."
# OpenAI
export OPENAI_API_KEY="sk-..."
# DeepSeek
export DEEPSEEK_API_KEY="sk-..."
# Venice AI
export VENICE_API_KEY="..."Note: If no API keys are set, SentinelAI automatically uses its built-in offline Heuristic Rule Engine with zero configuration.
pytest -vSentinelAI provides a REST API that powers the web dashboard and integrates into automated DevSecOps pipelines:
| Method | Endpoint | Description |
|---|---|---|
GET |
/api/v1/recon/{domain} |
Instant OSINT, DNS resolution, open port scanning, and WAF identification |
GET |
/api/v1/tools |
Curated security tools inventory with local host installation detection |
POST |
/api/v1/fuzz |
Fast sensitive path & file fuzzer with soft-404 wildcard filtering |
POST |
/api/v1/quick-scan |
Asynchronous standalone scan with zero database dependencies |
GET |
/api/v1/reports |
Index of all generated HTML, Markdown, and JSON audit reports |
GET |
/api/v1/reports/{filename} |
Interactive view or download of generated security reports |
POST |
/api/v1/scan |
Distributed cluster scan (uses PostgreSQL + Celery workers) |
GET |
/api/v1/scan/{id}/status |
Real-time status polling for distributed scans |
If you wish to run the full distributed cluster with PostgreSQL, Redis, Celery, and the React Web Dashboard:
cp .env.example .env
docker compose up -dAccess the web dashboard at: http://localhost:3000
FastAPI Backend documentation: http://localhost:8000/docs
SentinelAI is developed for authorized penetration testing, security auditing, vulnerability assessment, and educational research purposes only. Do not scan targets without prior written authorization from the system owners.
This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.