WAF + JWT.

A full-stack security application combining traditional WAF controls with machine-learning-based attack detection, JWT authentication, and LIME-powered explainability.

The WAF inspects incoming requests through security middleware, applies traditional controls such as IP blocking and rate limiting, and analyzes suspicious payloads using a Python machine-learning component. TF-IDF converts payload text into numerical features and a Logistic Regression model classifies requests as normal, SQL injection, XSS, path traversal, or command injection. LIME provides human-readable explanations for individual predictions.