This year marks three years since we published our CTU-SME-11 dataset: 11 devices, 7 days, 99 million expert-labeled network flows with real benign and malicious traffic. 🔗 Explore the dataset: doi.org/10.5281/zeno...
Stratosphere Laboratory
@stratosphereips.bsky.social
Researching at the intersection of Machine Learning, Cybersecurity and helping others. AI Center, Faculty of Electrical Engineering, Czech Technical University in Prague
We believe practical cybersecurity education should be available to anyone with curiosity and an internet connection, regardless of where they live or what they can afford. Help us spread the word! Learn with us through our free or professional track: cybersecurity.bsy.fel.cvut.cz
Registrations are OPEN for the “Introduction to Security” course 2026. 🔗 Read more:
Security of an In-orbit Satellite: Detection of Compromise Through Integrity 🔗 Read more:
Security of an In-orbit Satellite: Detection of Compromise Through Integrity — Stratosphere Laboratory
Small satellites are increasingly vulnerable to cyberattacks, yet their resource constraints make implementing robust security mechanisms a significant challenge. This thesis explores how to protect the...
stratosphereips.org
Poster: Multi-Objective Model Selection Pipeline for Network Flow Classification at POSTERS 2026 🔗 Read more:
Poster: Multi-Objective Model Selection Pipeline for Network Flow Classification at POSTERS 2026 — Stratosphere Laboratory
Training classifiers for network intrusion detection is hindered by two types of problems: data challenges (lack of labels, class imbalance, non-IID data, and concept drift) and engineering challenges...
stratosphereips.org
Dean's Award Outstanding Teaching "Lecturer Category" to Sebastian Garcia 🔗 Read more:
Dean's Award Outstanding Teaching "Lecturer Category" to Sebastian Garcia — Stratosphere Laboratory
Sebastian has just been awarded by the Dean of the Faculty of Electrical Engineering, Czech Technical University in Prague, for his outstanding teaching performance in the Winter Semester 2025/2026!
stratosphereips.org
NetSecGame v0.2.0 - Reproducible Experiments and a More Robust Game Server 🔗 Read more:
NetSecGame version 0.2.0 — Stratosphere Laboratory
NetSecGame v0.2.0 is here. This release focuses on what matters most for reproducible AI research: deterministic episode control, a more robust simulation server, and a significantly expanded test suite....
stratosphereips.org
NetSecGame - A Framework for Training and Evaluating AI Agents in Network Security Environments 🔗 Read more: www.stratosphereips.org/blog/2026/1/...
Build Cyber Agents with NetSecGame v0.1.0 — Stratosphere Laboratory
Build, train, and evaluate network security agents using NetSecGame. A high-speed, Docker-ready framework for Reinforcement Learning and cyber research.
stratosphereips.org
Adaptive Response in Slips IDS as Immune T Cells 🔗 Read more:
Adaptive Response in Slips IDS as Immune T Cells — Stratosphere Laboratory
The T Cell module was created to give Slips a stateful adaptive response layer on top of its existing evidence pipeline.
stratosphereips.org
Adapting Detections in Slips with Immune Pseudo-Generated Regexes 🔗 Read more:
Adapting Detections in Slips with Immune Pseudo-Generated Regexes — Stratosphere Laboratory
The RegexGenerator module was created to give Slips an adaptive way to discover new string-based detectors for changing indicators such as domains, URIs, filenames, TLS SNI values, and certificate common...
stratosphereips.org
HTTPS Anomaly Detection in Slips: Adaptive Baselines for Real Traffic 🔗 Read more:
New Slips IDS HTTPS Anomaly Detection Module — Stratosphere Laboratory
The new HTTPS anomaly detection module in Slips builds per-host adaptive baselines in traffic time, then detects deviations at two levels.
stratosphereips.org
What if cybersecurity worked more like the immune system? Not perfect prevention, but detection, response, learning, and adaptation over time. We wrote some thoughts on it! 🔗 Read more: www.stratosphereips.org/blog/2026/2/...
Rethinking Cybersecurity Defense: Principles from Biological Immunity — Stratosphere Laboratory
Our research identifies sixteen fundamental principles of biological immunity and translates them into cybersecurity defense architectures that emphasize multi-dimensional coordination over single-…
stratosphereips.org
NetSecGame - A Framework for Training and Evaluating AI Agents in Network Security Environments 🔗 Read more:
Build Cyber Agents with NetSecGame v0.1.0 — Stratosphere Laboratory
Build, train, and evaluate network security agents using NetSecGame. A high-speed, Docker-ready framework for Reinforcement Learning and cyber research.
stratosphereips.org
What if we told you we built a system that can break every LLM assistant you throw at it? Our tool presented today at #BHEU Arsenal, achieves just that! 💾 Explore more here: github.com/stratosphere...
The StratoCyberLab (SCL) is our local cyber range, where anyone can test and practice their offensive and defensive cybersecurity skills: github.com/stratosphere... 🔒 Local-only. No cloud. No tracking. No login. No data collection. ⚙️ Easy-To-Use. Only Docker is required. #CyberRange #Education
Woohoo! A new edition of our Stratosphere newsletter is just out! Catch up with some of our team activities since the last edition! 🚀 www.linkedin.com/pulse/strato...
If you missed our talk at #ekoparty, here's a poster summarizing some of our results on evaluating ARACNE, our multi-agent framework for autonomous Linux shell security testing. Our evaluation is still ongoing, and we hope to share more results soon 🚀
Every year, thousands join our CTU Introduction to Security — where curiosity meets courage, and learning means attacking & defending. This is our heartbeat. 🎬 Watch the premiere → www.youtube.com/watch?v=FLtX... #CyberSecurity #Animation
Official Intro Theme — CTU Introduction to Security
Every year, hundreds of students start Introduction to Security — a 14-week, hands-on course where curiosity meets courage, and learning means breaking, building, and defending…
youtube.com
🚀 At #ekoparty, our team presented ARACNE, a multi-agent LLM based offensive and defensive agent for Linux shell systems, and one of the largest LLM model evaluations to date. 🔗 Slides: bit.ly/EKO21ARACNES... 🔗 Code: bit.ly/EKO21ARACNEC... 🔗 Paper: bit.ly/EKO21ARACNEP...
AI Attackers in Your Pocket: How Small Language Models Can Outsmart Cyber Defenses 🔗 Read more:
AI Attackers in Your Pocket: How Small Language Models Can Outsmart Cyber Defenses — Stratosphere Laboratory
We are pleased to announce the publication of our latest paper, “ Building adaptive and transparent cyber agents with local language models ,” in the Journal of Expert Systems with Applications.
stratosphereips.org
Our colleague Lukáš Forst talked with Česká televize on the partial end of support for Windows 10, and what are the implications for European users. Watch the full segment in Czech here: www.ceskatelevize.cz/porady/10971...
