Show HN: Wiredigg – Real-Time Network Analysis with ML and Ollama Support

[justvugg]

I built Wiredigg, an open-source network traffic analysis and security tool written in Python. It combines real-time packet capture, protocol inspection, machine learning-based anomaly detection, and local LLM analysis via Ollama.

The goal is to provide interactive network visibility with AI-assisted threat interpretation, while remaining local-first and easy to run. A Windows executable build is also available.

Repo: https://github.com/JustVugg/Wiredigg

⸻

What it does

Real-time packet capture • Live traffic sniffing • Protocol analysis (TCP, UDP, ICMP, HTTP, etc.) • Filtering by IP, port, and protocol • Promiscuous mode support

Machine learning anomaly detection • Detection of unusual traffic patterns • Threat classification with severity levels • False-positive marking and incremental retraining • User-assisted model refinement

Ollama integration (local LLM) • Sends flagged anomalies to a locally running model via Ollama • Generates contextual, human-readable explanations • Adds reasoning on top of statistical detections • Fully offline if Ollama is local

Threat intelligence & dashboards • Malicious IP/domain checks • Interactive tables and traffic statistics • Graph-based visualizations • Exportable reports (HTML, JSON, text)

IoT & device analysis • Device identification and classification • Behavioral pattern analysis • Risk evaluation based on traffic activity

Custom packet tools • Manual packet crafting for testing • Control over IP, port, protocol, and payload

Running it

From source:

git clone https://github.com/JustVugg/Wiredigg pip install -r requirements.txt python wiredigg.py

Administrator/root privileges are required for packet capture.

Alternatively, you can use the provided Windows .exe build.

⸻

Why I built it

Many packet analyzers are either: • CLI-heavy and technical • Or large enterprise systems

I wanted something: • Visual • AI-augmented • Local-first • Extensible in Python • Usable for labs, small networks, and learning

Happy to get feedback, especially on the ML and Ollama integration approach.