AI‑powered middleware for server optimization and anomaly detection.
turboSH is an intelligent middleware layer that sits between clients and backend servers. It automatically:
- Schedules and rate-limits incoming requests to prevent overload
- Caches frequent responses to reduce backend load
- Logs traffic and extracts behavioral features
- Detects anomalies using machine learning (Isolation Forest, One‑Class SVM, LOF)
- Mitigates threats automatically (block, throttle, or allow)
Designed to run on commodity hardware — no GPU required.
You can run turboSH in front of any existing API without installing Go or Python. Make sure to build the image locally first!
docker build -t turbosh-proxy .Note: Replace host.docker.internal:9090 with the actual reachable URL of your backend.
docker run -p 8080:8080 -e TURBOSH_BACKEND="http://host.docker.internal:9090" turbosh-proxyWant to customize rate limits, ML thresholds, or integrate with Grafana?
👉 Read the official turboSH Playbook
Client → Reverse Proxy → Scheduler → Cache → Traffic Logger
↓
Backend ← Decision ← ML Inference ← Feature Extraction
See ARCHITECTURE.md for full details.
turboSH/
├── core/ Middleware components (proxy, scheduler, cache, security, decision)
├── pipeline/ Data pipeline (logging, feature extraction, dataset builder)
├── ml/ ML model training and evaluation
├── models/ Trained model artifacts (.onnx)
├── monitoring/ Prometheus metrics and dashboards
├── datasets/ Generated traffic datasets
├── notebooks/ Jupyter notebooks for analysis
└── docs/ Project documentation
| Layer | Technology |
|---|---|
| Middleware | Go (net/http, gin) |
| Data Pipeline | Go + Python (pandas, numpy) |
| ML | Python (scikit-learn, ONNX) |
| Monitoring | Prometheus + Grafana |
- Go: 1.24+
- Python 3.10+
# Clone
git clone https://github.com/Keshav76315/turboSH.git
cd turboSH
# Go dependencies
go mod tidy
# Python environment
python -m venv .venv
.venv/Scripts/Activate # Windows
# source .venv/bin/activate # Linux/Mac
pip install -r requirements.txt| Developer | Role |
|---|---|
| Keshav | Backend systems + ML engineering |
| Anzal | Data pipeline + caching system + data science |
| Document | Description |
|---|---|
| PLAN.md | Development plan (EPICs & stories) |
| ARCHITECTURE.md | System architecture & interfaces |
| PROGRESS.md | Development history |
| AGENT.md | AI agent context |
| DATA_SCHEMA.md | Log & feature schemas |
| API.md | Internal API definitions |
TBD