Tootfinder

We had difficulties automatically classifying chats, messaging channels, and forums. So we tested several open-weight large language models for our use case. The benchmark is published below, along with a tool supporting the classification process.
This helps us avoid manually classifying channels collected from dark-web forums and social networks, while reliably identifying which channels can be discarded and which deserve further analysis.
🔗 Publication #ai #llm #cybersecurity #darkweb #osint #opensource

I just released ptrclassify is a small, dependency-free Python library and CLI that infers likely IP usage from reverse-DNS PTR hostnames.
It is intentionally heuristic and multi-label. PTR naming is operator-controlled and is not authoritative evidence of how an address is actually used. The output therefore includes a confidence score, the text that matched, and the rule IDs that produced each label.
To summarize, the library is trying to guess usage (and a bit location) of an IP address based on its PTR records. It's based on a set of rules which can be updated easily.
#osint #cybersecurity #ptrclassify #opensource #dns
https://github.com/adulau/ptrclassify
module https://pypi.org/project/ptrclassify/