nfstream/nfstream

Health v2 · maintenance only

Adoption not part of the score

1218 stars · 145 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NFStream is a multiplatform Python framework for fast, flexible network flow data analysis from live interfaces or pcap files. It provides flow metering, nDPI-based encrypted layer-7 identification, statistical feature extraction, and extensible NFPlugin support aimed at reproducible machine learning research.

Use cases

• extract network flows from a large pcap file into pandas or csv
• identify encrypted applications in live traffic with nDPI
• extract statistical flow features for machine learning traffic classification
• capture flows from a network interface in python
• add custom flow features with a plugin
• generate reproducible network traffic datasets for research
• fingerprint TLS and SSH metadata from traffic

When to choose

• you need fast python-based flow aggregation from pcaps or live interfaces
• you want layer-7 application identification including encrypted traffic
• you are building ML models for network traffic management and need reproducible features
• you need extensible flow feature extraction via plugins

When to avoid

• you need full packet-level inspection rather than flow-level analysis
• you need a GUI packet analyzer like Wireshark
• you require a permissive license for proprietary linking (LGPL-3.0)
• you need non-python tooling

Facets

framework · maturity stable

nlp machine-learning data-science etl analytics developer-tools networking security data-science machine-learning analytics developer-tools python windows cross-platform network-analysis pcap netflow deep-packet-inspection ndpi traffic-classification packet-capture network-monitoring flow-features nfplugin linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="nfstream/nfstream")