Smart India Hackathon (SIH) Problem Statement ID: SIH26155
Theme: Cybersecurity / AI & Smart Automation
Category: Software
NetAuditAI is an intelligent, automated network security compliance auditing platform that ingests heterogeneous multi-vendor network device configurations (Cisco, Juniper, Palo Alto, Fortinet, CheckPoint) and audits them against global compliance frameworks (NIST 800-53, CIS Benchmarks, PCI-DSS, ISO 27001, HIPAA, SOC 2, GDPR, DPDP Act 2023) using RAG-powered neural analysis.
-
🔌 Multi-Vendor Ingestion & Auto-Detection:
- Ingests raw .txt, .conf, .cfg, .log configuration exports.
- Heuristic pattern-matching auto-detects vendor syntax (Cisco IOS, Cisco NX-OS, Juniper JunOS, Palo Alto PAN-OS, Fortinet FortiOS).
- Multi-device batch upload support.
-
📜 Comprehensive Framework Mapping:
- Multi-standard support: NIST 800-53, CIS Benchmarks, PCI-DSS 4.0, ISO 27001, HIPAA, and India's DPDP Act 2023.
- Flexible scoping: audit individual devices or apply global frameworks network-wide.
-
🧠 RAG-Powered AI Compliance Engine:
- Normalized intermediate configuration representation (AST-like JSON schema).
- ChromaDB semantic search maps configuration directives to compliance control requirements.
- Delivers deterministic PASS, FAIL, or WARNING verdicts with exact line-number evidence.
-
⚡ Actionable Remediation Generator:
- Automatically synthesizes vendor-specific CLI configuration snippets to fix failed controls.
- Dual-mode: CLI syntax rollback stanzas and best-practice hardening guidelines.
-
🗺️ Interactive Network Topology & Visualizer:
- Interactive node-link network visualization powered by ReactFlow.
- Visual device status highlighting compliant vs. vulnerable nodes.
-
📊 Executive Analytics & PDF Export:
- Interactive compliance posture dashboards with Recharts.
- One-click executive PDF report generation via @react-pdf/renderer.
| Component | Technologies |
|---|---|
| Backend API | FastAPI, Pydantic v2, Uvicorn, Python 3.11+ |
| Database & Cache | SQLAlchemy 2.0 (Async), SQLite / PostgreSQL, Redis, Celery |
| Vector Engine & RAG | ChromaDB, NetworkX, LangChain / LLaMA model adapter |
| Frontend SPA | React 18, Vite 5, TypeScript 5.2, TailwindCSS 3 |
| State & Fetching | Zustand, TanStack React Query |
| Visuals & PDF | ReactFlow (topology), Recharts (charts), @react-pdf/renderer |
| DevOps | Docker, Docker Compose, Nginx |
�ash git clone https://github.com/pravinsingh1109/NetAuditAI.git cd NetAuditAI
�ash docker-compose up --build
- Frontend App: http://localhost:5173
- Backend API: http://localhost:8000
- Interactive Swagger Docs: http://localhost:8000/docs
`�ash cd backend python -m venv venv
.\venv\Scripts\activate
pip install -r requirements.txt uvicorn app.main:app --reload --port 8000 `
�ash cd frontend npm install npm run dev
├── Docs/ # PRD, architecture, API, and design documentation ├── backend/ # FastAPI backend services & parsers │ ├── app/ # Core routing, models, schemas, and AI RAG engine │ ├── tests/ # Backend test suites │ └── requirements.txt # Python dependencies ├── frontend/ # React + Vite frontend application │ ├── src/ # React components, pages, stores, and hooks │ └── package.json # Node dependencies ├── samples/ # Sample network configuration files for testing ├── docker-compose.yml # Multi-container orchestration specification └── .env.example # Sample environment variables template
Pravin Singh
- GitHub: @pravinsingh1109
This project is open-source and licensed under the MIT License.