CIRCL Open Data at CIRCL

Open Data at CIRCL

Open cybersecurity resources

Data and models built to be reused

CIRCL publishes datasets, machine-learning models and operational security data so researchers, defenders and developers can study threats, reproduce results and build new tools.

Explore CIRCL on Hugging Face Browse the catalogue
Open by design
  • Use
  • Study
  • Adapt
  • Share

Why open data?

Security improves when knowledge travels

CIRCL advocates responsible data sharing. Making useful material available can lead to new research, independent analysis, software and services that improve security over the long term—both in Luxembourg and abroad.

We follow the Open Definition: open data and content can be freely used, modified and shared by anyone for any purpose, subject to the licence attached to each resource.

Featured platform

CIRCL on Hugging Face

Our Hugging Face organisation brings machine-learning resources into one discoverable, versioned hub. Always consult the repository card for a resource's latest documentation, files and licence.

Datasets

Explore published datasets

Browse data prepared for cybersecurity research, experimentation and reproducible analysis.

View all datasets

Models

Use CIRCL models

Find models released by CIRCL for security research and threat-analysis workflows.

View all models
Looking for everything in one place?Visit the CIRCL organisation profile

Data catalogue

More open resources

Alongside Hugging Face, CIRCL operates public services and direct downloads covering vulnerabilities, network reputation and historic malware observations.

Daily JSON exports

Vulnerability information

Daily dumps of CVE records and vulnerability information, separated by source, from Vulnerability-Lookup.

Browse vulnerability dumps
Query service

BGP Ranking

Historical rankings of malicious activity observed per BGP autonomous system number, available through the public service.

Open BGP Ranking
Image dataset

Phishing screenshots

More than 400 screenshots of verified or potential phishing websites, with supporting documentation.

View dataset details
Image dataset

Dark-web screenshots

More than 37,000 classified screenshots of dark-web websites collected for image-matching research.

View dataset details
Historic statistics

Allaple observations

Raw 2015 honeypot statistics for the Allaple malware family, published with CIRCL's technical report.

Read the report
Activity data

Operational statistics

Statistics covering CIRCL's incident reports, analysis requests, support activity and third-party notifications.

View operational statistics

Research

Image-matching publications

The image datasets support a wider body of CIRCL work on classification, evaluation frameworks and practical matching libraries.

Dataset publicationOpen data for image matching Classification toolVisJS-Classificator Evaluation frameworkCarl-Hauser Matching libraryDouglas-Quaid

Used CIRCL data in your research? Tell us about your work so we can learn from it and consider featuring it here.

Reuse responsibly

Licensing and classification

This page is TLP:CLEAR and may be distributed without restriction. Unless a repository or dataset states otherwise, the open data described here is licensed under the Creative Commons Attribution 4.0 International licence.

Before using a resource, check its own documentation for the authoritative licence, scope, provenance and any handling guidance.

Top