NetSentinel AI

Autonomous Network Intelligence — Not just a config tool, an autonomous AI agent that plans, executes, verifies, and documents network changes like a senior engineer.

Built at HackFax 2026.

---

What It Does

NetSentinel AI is a full-stack platform that brings autonomous AI agents to network management. Instead of manually writing CLI commands, troubleshooting device-by-device, or generating configs from templates — you describe your intent in plain English, and NetSentinel handles the rest.

Core Agents

Agent	What It Does
Orchestrator	Breaks down high-level goals into multi-step executable plans with risk assessment
Troubleshooter	Diagnoses network issues — forms hypotheses, runs diagnostics, identifies root cause, proposes fixes
Config Generator	Converts plain-English intent into vendor-specific CLI configs with validation and rollback procedures
Chat Assistant	Conversational AI for general network Q&A and guidance

Key Features

• Intent-Based Configuration — Describe what you want ("Add VLAN 40 for Finance on all access switches") and get production-ready configs
• AI-Powered Diagnostics — Input a symptom, get ranked hypotheses, diagnostic steps, root cause analysis, and a proposed fix
• Autonomous Plan Execution — Multi-step plans with discovery, validation, configuration, and verification phases
• Real-Time Terminal View — Watch agent execution with streamed CLI output via Server-Sent Events
• Network Inventory Dashboard — Monitor device status, view configs, connected assets, and security logs
• Full Audit Trail — Every autonomous action is logged with timestamps, agent source, and affected devices
• Rollback Support — Every change is tracked with rollback procedures ready to execute

---

Tech Stack

Layer	Technology
Frontend	React 19, TypeScript, Vite, Tailwind CSS
Backend	Python, FastAPI, SQLAlchemy, SQLite
AI	Google Gemini 2.0 Flash with structured JSON output
Streaming	Server-Sent Events (SSE) for real-time execution

---

Run Locally

Prerequisites: Node.js, Python 3.9+

Frontend

npm install
npm run dev

Backend

cd backend
pip install -r requirements.txt

Create backend/.env with your Gemini API key:

GEMINI_API_KEY=your_key_here

Start the server:

python -m uvicorn main:app --port 8000 --reload

API Endpoints

Method	Endpoint	Purpose
POST	/api/plan	Generate orchestration plan from a goal
POST	/api/troubleshoot	Diagnose a network symptom
POST	/api/config	Generate config from intent
POST	/api/chat	Chat with the AI assistant
POST	/api/execute	Stream device execution (SSE)
GET	/api/logs	Retrieve audit logs
GET	/api/changes	Get change records
POST	/api/changes/{id}/rollback	Rollback a change
GET	/api/health	Health check