Tejas Raksha: Zero-Day Scanner for Agriculture Web Portals

Modern agricultural logistical networks, state welfare systems, and public commodity marketplaces rely heavily on public web portals to process citizen transaction logs, land distribution records, and sensitive identity files. Despite their critical importance, these specialized public networks have faced an aggressive 300% increase in automated web attacks over a two-year window. Traditional scanning tools are either too expensive or create massive volumes of false alarms that cause severe alert fatigue. To bridge this gap, this study introduces Tejas Raksha (Sanskrit for "Protection of Light"), an affordable, open-source, and multi-tenancy-ready asynchronous security scanner engineered specifically for resource-constrained e-governance systems. The platform operates as a distributed microservices mesh spread across six independent, containerized layers managed via Docker Compose: a React 18 frontend, a FastAPI gateway backend, a Celery task worker, a Celery Beat periodic scheduler, a PostgreSQL database, and a Redis message broker. The scanner utilizes an asynchronous crawl pattern to autonomously test for 12 distinct categories of OWASP Top 10 vulnerabilities across a configurable crawl depth (1–10) and concurrency limit (1–50) while fully respecting robots.txt directives. Crucially, the framework addresses alert fatigue by introducing an intelligent, Three-Layer False Positive Verification Engine that runs Content Signature Verification, Dynamic Error Baseline 404 Fingerprinting, and String Exclusion Matching Matrices. Empirical load tests with Locust demonstrate high-throughput stability at 850 concurrent requests per second, sub-100 ms API response times, and professional automated PDF report generation speeds under 5 seconds. Real-world validation sweeps against standardized vulnerable deployment targets confirm an average 94% vulnerability detection confidence rating while systematically compressing traditional background tool noise down to a precise ~15% margin.