BSidesVienna 0x7EA

TLDR: If you as an attacker want more tools to gain RCE and persistence MCP is exactly that.

AI agents are rapidly becoming a new interface to enterprise systems: they read internal knowledge, call APIs, and execute actions through connected tools. MCP standardizes this tool access, but it also creates a new, high-impact attack surface: tool execution integrity.

In a lab environment we are going to do some hands-on sessions about NTLM-Relaying and showing common protection mechanisms that are effective against these kinds of attacks. So the workshop should be interesting for sysadmins and pentesters alike.
It ties into our talk about modern NTLM-Relaying methods and lets participants try out the attacks in a simulated environment provided by us.

AI has industrialized cybercrime. Exploit timelines collapsed from days to hours, ransomware victims surged 389%, and 4.6 billion stolen credentials flooded darknet markets. Then Claude Mythos showed autonomous zero-day discovery at scale. This talk presents current threat data, examines how AI reshapes both offense and defense, and offers practical strategies for security leaders.

This talk is about SBOM — the thing that has become the most important compliance artifact in software development. I'll cover:

• Why? A brief, painful tour through the legal plot twists that turned the SBOM from a nice-to-have into a "you literally cannot sell this product without one" because too many discovered they had no idea what they were actually running.
• What? what actually goes in one, what people think goes in one, why legal and IT security want one, and what happens if you vibe-coded the whole thing.
• Who? What this all means for software developers, open source maintainers who never signed up to be vendors, and where the liability actually lies.

Expect war stories, regulatory translation services, a healthy amount of "I am not your lawyer, but…", less “it depends” and a takeaway list of things you can actually do on Monday morning — whether you're shipping software, consuming it, or maintaining the one library that all of your products depend on.

"Friends don't let friends upload their tradecraft to VirusTotal", but what about AI? More and more companies are integrating AI into their pipelines and workflows, and we can see headlines of AI finding hundreds of bugs everywhere online. So how can we integrate this technology into our work without burning our tradecraft - is this even possible?

Possibly you already know lock picking. Possibly not.
We explain what happens in the lock while picking or regularly with a key.
Lock picking itself, we will show you in an own workshop at the bsides vienna.

Additional, there are competitions. For example the yearly austrian championship.
Or the belt-system, which be presented.

Mobile app testing has many pitfalls and a structured approach is needed to get a holistic picture of the attack surface. The OWASP Mobile Application Security (MAS) project is able to support you with that. In this talk, you'll get a practical introduction to the MAS ecosystem which consists of the standard, mobile weaknesses and how to test them. You will see test cases and demos for iOS and Android in action for static and dynamic analysis, and learn how to perform a mobile penetration test on a non-jailbroken iOS device.

Using Beacon Object Files (BOFs) to execute external post-exploitation capabilities from a C2 agent has been a staple technique in offensive security for years now. The self-contained programs are great for one-off tasks, such as gaining situational awareness, elevating privileges or dumping credentials. Recently, BOFs have received a powerful upgrade that allows for them to be executed in the background, enabling long-running real-time monitoring functionality. This talk shows how the Conquest framework supports Async BOFs and how they can benefit modern red teaming.

For years, the cybersecurity community has worshiped at the altar of the MITRE ATT&CK® framework to track digital adversary behaviors. We spend millions hardening the cloud, fine-tuning EDRs, and chasing zero-days, yet we consistently forget one glaring reality: an adversary can just walk through the front door with a high-vis vest and a clipboard.
Enter PACT (Physical Access & Control Taxonomy). Built to bridge the grand canyon between meatspace and cyberspace, PACT is an open-source, community-driven framework that translates physical tactics, techniques, and procedures (TTPs) into a structured matrix that mirrors the MITRE ATT&CK format.

Really short: You can try to open a lock without a key for you first time.
You do not need own tools for that, we provide everything you need, also the locks.
If you have your own tools, you can use it.

After the workshop, you would be ready for the Austrian championship next year.

HackerOne Club Austria!

This club is dedicated to building a collaborative and welcoming hacker community across Austria. Whether you are an experienced bug bounty hunter, cybersecurity professional, student, or someone just getting started in ethical hacking, you are welcome to join and grow with the community.

The Austria HackerOne Club organizes a wide range of events including:
- local meetups
- live hacking sessions
- educational workshops
- Capture The Flag (CTF) competitions
- conference gatherings
- world cup meetings (AWC)
- discussions focused on bug bounty, ethical hacking, and cybersecurity.

In Cybersecurity there is a narrative existing: The problem is sitting in front of the screen… but is it?

Our understanding of humans in cybersecurity is shaped by problematic metaphors, which influence how we design security systems. The way we likely describe humans, shapes how we approach cybersecurity.

Humans are as seen as the weakest link: Humans are viewed as the main source of failure > Assumption: Technology is strong, humans are weak

Humans are seen as driven by fear: as frightened animal > Assumption: Fear and punishment drive secure behavior

And once you believe, that the human is the problem, you stop looking for better explanations.

In our talk we will have a deeper look at these assumptions and the psychological as well as technical factors of (in)secure behavior in organizations
Cognitive biases often cause individuals to underestimate rare but catastrophic risks or to place excessive trust in automation. Routine blindness may result in subtle anomalies being ignored when tasks become repetitive. Furthermore, poor collaboration and information silos weaken collective intelligence, while misguided prioritization—such as choosing convenience over security—can undermine defense efforts. Yet, to fully leverage the strengths like pattern recognition, intuition, adaptive reasoning and ethical decision making, organizations should minimize human error through training, supportive tools, and sustainable working conditions, ensuring that human intelligence can function as a powerful ally in defending against digital threats.

Net Shredder is a network coverage-guided fuzzer for the Linux kernel. Built with a modular approach, and focused on the ease of use, it could be adapted to fuzz other types of targets (e.g., usermode applications) as well. During development, it found three remote vulnerabilities in the Linux kernel, one of which resulted in CVE-2025-22037.

This talk explores offensive security techniques against ML and LLM systems, from adversarial inputs and data poisoning to prompt injection and model extraction, with a live local demonstration of indirect prompt injection against a self-hosted model. The attendees will learn about the risks of deploying AI systems without proper security measures.

How drones (UAV’s) are the threat everyone saw coming but no one acted in time. Drones are the "Gray Rhino" of modern security: a visible, high-probability threat we failed to mitigate. This talk explores how rapid UAV proliferation outpaced regulation, creating critical vulnerabilities in privacy and infrastructure. We’ll analyze the shift from hobbyist novelty to sophisticated aerial risk and the urgent need for counter-UAS strategies.

Four years ago, Project TEMPA exposed the fundamental vulnerabilities within Tesla’s Bluetooth Low Energy (BLE) Passive Entry system and the Vehicle Controller Secondary (VCSEC) protocol. The research demonstrated how trivial it was to execute Man-in-the-Middle (MitM) relay attacks to unlock and drive away modern vehicles. In response, Tesla embarked on a multi-year effort to overhaul its access systems. But did they actually fix the underlying problems, or just change the locks?
​In this presentation, we are getting "phonkey" again. We will dive into a comprehensive 4-year retrospective of Tesla’s Phonekey security evolution, dissecting the ongoing tug-of-war between seamless user experience and robust vehicle security.

This talk examines how established macOS exploitation techniques can be applied to a largely overlooked attack surface: audio plugin installers. By analyzing installers from multiple major vendors, I demonstrate how common design and implementation flaws can be leveraged to achieve local privilege escalation. The presentation covers nine CVEs across five different vendors, highlighting recurring vulnerability patterns, exploitation strategies, and the security implications for both developers and end users. Attendees will gain insight into the intersection of macOS installer security and the audio software ecosystem, along with practical lessons for identifying and mitigating similar issues.

"Secrecy is not a technical problem with a technical solution — it is a social contract so old we've forgotten we signed it."
Before encryption. Before firewalls. Before the first wax seal was pressed into a letter — someone decided that some things should not be known by everyone. Why?
This workshop doesn't start with algorithms. It starts much earlier: with the question of why humans began keeping secrets at all, what social and cognitive machinery that required, and what it might mean to dismantle those categories entirely.

New versions are usually exciting and full of possibilities, but keeping up with every new edge-case can be a very exhausting task. Especially on Android tools go out of date quickly and the latest research paper's implementation probably relies on a Java version most developers roll their eyes at.

But binaries? Binary tooling is forever. Well, let's say slightly more stable. We'll explore on a practical example how to use the Android Runtime to compile apps' Dalvik bytecode into binary ELFs and use BinDiff for similarity analysis.

Active Directory remains the core identity system in most enterprise and governmental environments, making it a primary target for attackers after initial network compromise. Once inside a network, adversaries typically focus on AD reconnaissance, privilege escalation, and lateral movement in order to gain full domain control.

This presentation explores how attackers perform Active Directory enumeration using common tools and techniques, and why traditional security monitoring often fails to detect these early-stage activities. It then introduces deception-based defense strategies as an effective approach for early detection of malicious behavior within identity infrastructures.

The session focuses on the use of Active Directory honeypots and canary tokens as proactive detection mechanisms. These decoy assets are designed to appear legitimate within the environment while acting as high-fidelity tripwires for suspicious activity. Any interaction with these objects can immediately signal potential reconnaissance or compromise attempts.

Through practical examples and a simulated attack scenario, the talk demonstrates how deception techniques can detect attacker behavior during directory enumeration, credential discovery, and privilege mapping. The presentation also highlights how these mechanisms integrate into Purple Team methodologies and support incident response and forensic investigations.

Attendees will gain insight into how deception technologies enhance visibility within Active Directory environments, reduce attacker dwell time, and enable earlier detection of identity-based attacks before they escalate into full domain compromise.

Open Source Quantum Secure VPN - QKD + PQC over WireGuard

Quantum computers will break the encryption we rely on today, and the clock is ticking. But what does that actually mean, and what can you do about it right now?
This talk starts from the ground up, giving attendees an accessible introduction to the two quantum-safe building blocks: Post-Quantum Cryptography (PQC), a mathematical software-based replacement for classical key exchange, and Quantum Key Distribution (QKD), a physics-based approach that uses quantum optical channels to distribute keys with information-theoretic security. No prior knowledge of quantum mechanics required.
The second half of the talk is around ARNIKA (https://github.com/arnika-project/arnika) and how to build a fully open-source quantum secure VPN that wraps WireGuard with both PQC (Rosenpass) and QKD (via the ETSI014 API) into a single deployable solution.

Attendees will leave with:
QKD and PQC fundamentals, what they are, how they differ, and when to use each
A technical dive into open-source quantum secure VPN architecture, key derivation, ETSI014 integration, and WireGuard PSK injection

Target audience:
Security professionals and engineers curious about quantum-safe networking, no quantum physics or crypto background needed.

The answer to this question seems quite straightforward until we really start to think about it. My talk is inspired by a concept of Fred Brooks' 1986 paper "No Silver Bullet": the difference between essential and accidental complexity of software, only the first of which is an inherent and unremovable part of it. I want to investigate if such an "essence" of software really exists, and what follows if we take this view seriously.

This talk tells the story how a “let’s quickly look at the files” span to a multi-day journey into ancient Windows CE filesystems and NXP processor NAND flash handling in an embedded system. Expect a deeply technical talk that explains how handling single bits make a difference and lets you peek into the complexity below (embedded) operating systems.

In the last few years, I was responsible, together with my colleagues, for the fully open-source framework "Attackbed". We developed simulated networks with automated attacks and collected data for further evaluation of security measures or for building machine learning datasets. One example of such an evaluation is testing the performance of LLMs when used to detect malicious behavior in logfiles. We not only created a simulation with complex attack chains but also developed several public Metasploit Exploits and an automated attack tool. I will cover all the technologies and attack chains, and provide a perspective on potential use cases for this open-source framework.

In this talk, we walk through a real intrusion observed in an EDR-monitored enterprise environment. The case did not start with a major incident or a flood of alerts. It began with two ambiguous notifications in the Defender portal that the customer could not immediately classify. What looked like a minor signal turned into a live hunt: an operator attempting fileless execution, interacting with endpoint controls, trying to disable or bypass defenses, and carefully pivoting through the network.

CBOR (RFC 8949) is a binary serialization format used in constrained security-critical systems like FIDO2/WebAuthn and COSE. Despite a precise specification, implementations diverge across languages and systems, leading to different behavior when confronted with the same input. In this talk, 11 CBOR parsers across seven languages are compared to identify security-relevant behavior, such as unexpected acceptance/rejection of input, hangs and crashes.