PathHelm

Overview

Most API gateways are reactive - they block bad traffic after you write rules. PathHelm is proactive. It uses an IsolationForest machine learning model to detect anomalous traffic patterns in real-time, flagging suspicious requests before they become a problem.

The gateway includes all the standard features - rate limiting, IP blacklisting, circuit breaker patterns, and round-robin load balancing - but every component is configurable through a clean API. The real-time Streamlit dashboard lets you watch traffic flow, see anomaly scores, and adjust thresholds on the fly.

Under the hood, Redis handles session state and rate limit counters, FastAPI serves the gateway logic, and scikit-learn powers the anomaly detection. Everything runs in Docker containers for easy deployment.

Key Features