Grab your badge and freebies, then check out the CTF, exhibitors, lock pick village, and raffle table.
Grab your badge and freebies, then check out the CTF, exhibitors, lock pick village, and raffle table.
Grab your badge and freebies, then check out the CTF, exhibitors, lock pick village, and raffle table.
Grab your badge and freebies, then check out the CTF, exhibitors, lock pick village, and raffle table.
Grab your badge and freebies, then check out the CTF, exhibitors, lock pick village, and raffle table.
Live in Track 1, Simulcast in Tracks 2-3
Remarks will be delivered by BSidesAugusta staff and special guests.
Live in Track 1, Simulcast in Tracks 2-3
Remarks will be delivered by BSidesAugusta staff and special guests.
Live in Track 1, Simulcast in Tracks 2-3
Remarks will be delivered by BSidesAugusta staff and special guests.
Live in Track 1, Simulcast in Tracks 2-3
Remarks will be delivered by BSidesAugusta staff and special guests.
Live in Track 1, Simulcast in Tracks 2-3
Remarks will be delivered by BSidesAugusta staff and special guests.
Live in Track 1, Simulcast in all other tracks/rooms
Tim Kosiba, Deputy Director of the National Security Agency, will be our keynote speaker!
Live in Track 1, Simulcast in all other tracks/rooms
Live in Track 1, Simulcast in all other tracks/rooms
Tim Kosiba, Deputy Director of the National Security Agency, will be our keynote speaker!
Live in Track 1, Simulcast in all other tracks/rooms
Live in Track 1, Simulcast in all other tracks/rooms
Tim Kosiba, Deputy Director of the National Security Agency, will be our keynote speaker!
Live in Track 1, Simulcast in all other tracks/rooms
Live in Track 1, Simulcast in all other tracks/rooms
Tim Kosiba, Deputy Director of the National Security Agency, will be our keynote speaker!
Live in Track 1, Simulcast in all other tracks/rooms
Live in Track 1, Simulcast in all other tracks/rooms
Tim Kosiba, Deputy Director of the National Security Agency, will be our keynote speaker!
Live in Track 1, Simulcast in all other tracks/rooms
Memory analysis is an essential investigative technique for detecting modern, memory-resident malware. Unfortunately, it is often perceived as a difficult and manual process requiring expert knowledge of operating systems and malware internals. This talk challenges that assumption by presenting five checks that can be automatically performed against Windows memory samples to quickly surface signs of malicious code execution or attacker access. The presented checks answer five critical triage questions: 1) are any processes running with unexpected privileges, 2) are restricted resources being accessed by unexpected processes, 3) are any processes executing code in a suspicious manner, 4) has malware entered the kernel, and 5) does any process or kernel driver match known threat intelligence? All checks will be demonstrated using Volatility 3, the most widely used open source framework for memory analysis. Attendees will leave with a compact workflow that SOC analysts, threat hunters, and incident handlers can use to automatically identify memory-only malware, toolkits, and attacker activity, helping them rapidly prioritize systems during large-scale triage and decide where deeper forensic analysis is warranted.
Ai has taken over all of the defense discussion in the last couple of years. Will it help? Will it result in all of our jobs going away? If so how? In this talk we'll cut through all the hyperbole and focus on the data. What can AI do and no do and what does that mean for us as cyber-defenders? The future is incredibly interesting. We are on the cusp of a new era of cyber-defense and it's up to you to seize it. This talk is about how to do exactly that.
Chromium is no longer just a place to steal cookies from. It is a place for persistence and post exploitation for adversaries.
This talk starts with remotely enabling Chrome DevTools Protocol (CDP) inside a live Chrome or Edge process as a post-exploitation primitive. Once CDP is exposed, the browser’s own debugging interface becomes an operator console: enumerate or open new tabs, capture screenshots of active browser sessions, steal browser cookies, inspect a user's history, enumerate and load extensions, proxy authenticated requests, and interact with web apps from the user’s trusted browser context all in a hidden context from the user. Instead of fighting protections like device-bound session cookies, we will highlight how we can operate through the browser they are bound to.
Then we take the idea further: from browser control to browser persistence. We’ll examine how operators can abuse Chromium’s extension ecosystem, Native Messaging hosts, and Isolated Web Apps can be chained into a Mythic C2 agent for persistence. Along the way, we’ll cover how extension allowlist policies can be bypassed when trust is anchored to extension IDs.
This talk will also enable defenders by illuminating what artifacts are left behind when operators abuse chromium browsers for post-exploitation primitives.
This isn't your standard point-and-click exercise. The team at SteelGate LLC have been working overtime to spin up a custom playground of hackable web apps, pwnable services, and brain melting cryptography just waiting to be cracked wide open.
We are dropping you and your team into a Jeopardy style gauntlet. Whether you're a seasoned operator hunting for zero-days or just looking to pop your first shell, this environment is designed to test your limits and hopefully learn something new!
No advance registration is required. Just show up with your team(or just yourself), pick a handle, and get to hacking. A Wi-Fi enabled laptop loaded with your hacking suite is required to participate. We will have a few loaner laptops pre-loaded with Kali Linux for those that need one, available on a first come first served basis.
Check out https://sgctf.lol/ for updates.
Abstract
Configuring an Android penetration testing environment historically meant wrestling with broken $PATH variables, driver conflicts, complex SDK installations, root bypass workarounds, and strict administrative machine policies[cite: 1]. JAMBOREE changes all of that[cite: 1]. Designed to make mobile application security accessible to everyone, JAMBOREE is a portable, no-admin installer that provisions a fully functional Android security testing stack in seconds[cite: 1].
In this 4-hour hands-on workshop, intermediate security professionals will break down the myth that mobile testing requires hours of tedious setup and maintenance[cite: 1]. The first half of the session guides attendees through initializing a portable lab environment—spinning up a rooted Android Virtual Device (AVD) complete with Magisk, Burp Suite interception configurations, Java/Python runtime dependencies, and Objection—all without requiring administrative privileges on their Windows host machines[cite: 1].
In the second half, participants put the stack to work in real-time[cite: 1]. Attendees will intercept live HTTP/HTTPS traffic, dynamically inspect application runtimes, and analyze off-the-shelf, real-world Android applications[cite: 1]. If time permits, attendees will have the freedom to unpack and inspect target apps of their own choosing under guided supervision[cite: 1].
Prerequisites
- Knowledge: Basic understanding of Android architecture and core operating concepts[cite: 1]. Prior experience with device rooting, custom ROMs, or mobile modding is strongly recommended[cite: 1].
- Hardware: A Windows laptop with internet access (no local admin rights required—only WSL or "virtual machine platform" enabled on the laptop)[cite: 1].
Key Takeaways
- Spin up a complete, zero-admin Android security testing environment on Windows using JAMBOREE[cite: 1].
- Configure and operate a rooted Android emulator integrated with Burp Suite and Objection[cite: 1].
- Master dynamic analysis techniques for real-world Android applications in real-time[cite: 1].
NOTES: Attendance in this session will limited. First come, first served. We will break for lunch at regular lunch time.
Break
Break
Break
AI agents are showing up in every enterprise. They read email, browse the web, query internal systems, and take actions on their own. Leadership wants them deployed this quarter, and security owns the risk without a playbook. Prompt injection is the top risk on the OWASP Top 10 for LLM Applications, and it is the reason security teams cannot fully trust these agents. An agent cannot tell the difference between data and instructions. A malicious email, web page, or document can give an agent orders, and the agent will follow them using the credentials and permissions of whoever it works for.
We will walk through how these attacks work in plain English, using real cases: the EchoLeak zero click data exfiltration in Microsoft 365 Copilot and the ForcedLeak attack on Salesforce Agentforce. We will also cover why the common fixes fall short. Better system prompts, output filters, and newer models all help, but none of them stop the attack.
The second half is about what works. The same controls we already use for people apply to agents: least privilege, separation of duties, egress control, human approval for high risk actions, and audit logs. You will leave with a simple way to explain prompt injection to your CISO, a checklist for reviewing any agent deployment, and first steps for finding and monitoring the agents already running in your environment.
No AI background needed. If you understand phishing and insider threat, you already understand this talk.
Attacks against critical infrastructure organizations have increased from 2025 to 2026 including definitive proof of AI-enabled attacks. Claude and other tools are being leveraged for low sophisticated threat actors to speed up their payload development timeline. Mixed with the rise of hacktivist organizations there is a pervasive theme among these incidents. "This attack could have been prevented by basic security controls and defensible architecture".
Despite years of discussion around the cybersecurity talent shortage, organizations continue to struggle to fill entry level roles. Job postings demand experience that “entry level” candidates cannot realistically have, while capable, motivated learners are overlooked. The result isn’t just a pipeline problem it’s a misalignment between how talent is developed and how it is evaluated.
From the perspective of a high school cybersecurity educator actively preparing students for certifications, competitions, and early career pathways, this talk offers a grounded look at what the next generation of analysts actually brings to the table and where the gaps truly exist. These students are not starting from zero; many already possess foundational technical skills, hands-on lab experience, and a strong security mindset. However, they often lack exposure to professional workflows, communication expectations, and real-world operational context.
This session will break down the disconnect between classroom preparation and industry expectations, highlighting common misconceptions about entry-level talent. Attendees will gain insight into what emerging candidates can already do, what they still need to learn, and why traditional hiring filters may be excluding high-potential individuals.
Most importantly, this talk provides actionable strategies for bridging the gap redefining entry level expectations, designing meaningful early career opportunities, and engaging with talent earlier in the pipeline. Whether you are a hiring manager, team lead, or practitioner involved in mentoring, you will leave with practical ways to better identify, develop, and integrate the next generation of cybersecurity professionals.
Enjoy lunch provided by Chick-fil-A! A vegetarian option will also be available.
Lunch is served in the lobby of the Shaffer MacCartney Building.
Enjoy lunch provided by Chick-fil-A! A vegetarian option will also be available.
Lunch is served in the lobby of the Shaffer MacCartney Building.
Enjoy lunch provided by Chick-fil-A! A vegetarian option will also be available.
Lunch is served in the lobby of the Shaffer MacCartney Building.
Join us for an exciting, code-first exploration where we transform a humble chat completion script into a sophisticated, fully agentic AI system, all through a series of live, iterative Python demonstrations.
We'll kick off with a memory-less agent that does well with single questions but stumbles on conversational follow-ups. Watch as we add persistent message history, enabling coherent multi-turn dialogues. Next, we'll hit the wall of context limits when tackling file operations and large data, revealing why raw tool access isn't enough.
Through hands-on additions of tool-calling capabilities, we'll differentiate between built-in tools and configurable MCP services. Our agent will gain "hands" to act on the world, "eyes" to inspect complex data structures (databases, logs, packets, registries), and advanced memory architectures to manage its cognitive load.
Culminating in a blueprint for agentic memory management, including adaptive system prompts, modular skill loading, planning modes, and compressible context, we'll demonstrate how these elements combine to create truly autonomous, efficient AI agents.
After 9 years of hunting for malware, the ScumBots project has found and successfully analyzed over 23,000 malware binaries. It has also found over 10,000 binaries that it couldn’t recognize or analyze. So I did what any sane person would do, I trained a machine learning model to try and analyze the 10,000 unknown files. I’ll explain why I did it, how I did it, and what I learned along the way. I’ll also give you the tools to do this for yourself, and also give you the model files trained from ScumBots corpus of known malware.
A typical enterprise has hundreds to thousands of Chrome extensions running against its most sensitive systems, and little visibility into what those add-ons can actually do. An extension a single employee installs to "make Gmail nicer" may quietly hold permission to read cookies on every site the user visits. And "every site" includes your SSO, your secrets manager, your source control, your CRM.
We flipped the default: an allow-list posture where new extensions must be requested, and an evaluation engine that decides each request, acts on the decision, and explains it in plain English to the person who asked. This talk walks through how we built it and what we learned.
Evaluating a request means asking two different questions. The first is universal: a dangerous capability is dangerous no matter who installs it, and the engine checks for that like any scanner would. The second is the one a generic score structurally can't see — the environment the extension is running in. An extension that reads cookies across every site it can reach is a recognizable pattern anywhere. What an off-the-shelf tool can't do is resolve that "every site" against the specific hosts where a given org's SSO, secrets manager, and source control actually live, or know that the org's policy rules out a whole class of tool that's perfectly normal elsewhere. We'll dig into how the engine layers that context on top of the general evaluation, weighing capabilities in combination rather than in isolation and measuring access against an explicitly modeled set of the resources an org actually cares about, to land on an allow, block, or escalate-to-human decision.
The system spans both sides: the analysis that produces a verdict and the action taken on it. A user asks for an extension through the Chrome Web Store's own request flow; the request is enriched, evaluated, and acted on automatically — Chrome policy applied, a review ticket opened for the hard cases, the right people notified. The same model of the environment that produces the verdict also shapes the guardrail: rather than a coarse approve-or-block, Chrome's policy schema lets an extension be allowed but fenced off from the specific hosts that environment flagged as sensitive, and the engine drives that control straight from the verdict.
Automated security decisions tend to fail as black boxes: users route around controls they don't understand, and reviewers don't trust verdicts whose reasoning they can't see. So we treated the explanation as part of the control, not a log line. Alongside the decision, the engine assembles a plain-English justification from the facts that drove it, collapsing overlapping reasons into something that reads the way a human reviewer would write it. That justification surfaces in the portal: the interface where employees see what's approved, blocked, or under review, where a blocked extension becomes a self-service path to request a review rather than a dead-end, and where admins make and record the manual-review calls. The portal is where the whole system stops being a policy engine and becomes something people will actually use instead of work around.
We run all of this in production today on static analysis: we reason about what an extension declares it can do, not what it does when it runs. We'll talk honestly about where that line holds, and where we're taking it next — dynamic analysis of runtime behavior.
Attendees should leave able to reason about browser-extension risk in the context of their own environment, not a generic score built for someone else's.
This talk provides an attacker focused look at how adversaries can exploit these environments through techniques such as prompt injection, RAG poisoning, tool abuse, privilege escalation, and data exfiltration. Using real world attack scenarios, we will explore how individual weaknesses can combine into larger compromise paths that are difficult to detect with traditional security approaches.
The session will also discuss practical defensive strategies, including secure agent architecture, permission modeling, monitoring, and risk mitigation techniques. Attendees will gain a clearer understanding of how attackers approach multi-agent AI systems and how organizations can build more resilient AI deployments.
Ransomware negotiations are rarely just about money. They are high-pressure engagements shaped by psychology, leverage, timing, and uncertainty, all while organizations are still assessing the scope of compromise.
This talk offers a practitioner’s perspective on negotiating with threat actors during live extortion events. Drawing on real-world cases, it examines how threat actors communicate, what their behavior signals, and how disciplined negotiation strategy can influence outcomes beyond the final payment amount. Attendees will learn how pacing, message control, and delay are used to buy time, reduce risk, and preserve decision-making authority under crisis conditions.
The session also addresses common misconceptions about ransom negotiations, frequent mistakes that increase cost and exposure, and how negotiation fits into a broader incident response strategy that balances technical, legal, financial, and human factors. This presentation is about what works, and what doesn’t. What you should do, and maybe more importantly, what you should never do.
A real world Capture the Flag exercise that was designed for a human adversary. A new edition involving a red teaming AI agent. The agent is given a prompt representing the mission statement: to capture a set of 5 flags within a 30 minute time window. The defender’s goal? Impacting the agent and attempting to deny the agent from completing the CtF challenge.
This session is a dive into agentic attacks, why the dynamic reasoning and concurrency aspects of agentic attacks matter more than the machine-speed of execution, and the defender’s journey in designing countermeasures. The talk distills the essence of a range of CtF exercises performed with a team of agents and the learnings from the defender’s perspective.
In many organizations, Cyber Threat Intelligence (CTI) is viewed as a high-cost luxury reserved for mature SOCs. However, when implemented strategically, CTI is one of the most cost-effective force multipliers available to a security leader. By moving beyond simple "indicator feeds" and focusing on actionable context, organizations can simultaneously sharpen their technical response and revolutionize their security awareness culture.
This session explores a pragmatic, low-cost approach to building a CTI-driven security program. We will demonstrate how real-world threat data can be used to enrich the knowledge of security analysts, moving them from reactive alert-clearing to proactive threat hunting. Furthermore, we will show how to transform "boring" security awareness training into a dynamic, intelligence-led engagement program that uses current, industry-specific threats to educate employees.
Attendees will learn how to:
- Leverage Low-Cost Intelligence: Utilize OSINT, ISACs, and community sharing to build a high-value program on a budget.
- Enrich Technical Teams: Use adversary TTPs to prioritize patching, tune detection logic, and build analyst critical thinking skills.
- Modernize Awareness: Replace generic training templates with real-world intelligence that resonates with employees and turns them into active "human sensors."
- Speak to the Board: Translate technical threat data into the language of business risk and ROI to gain executive buy-in.
Whether you are a solo practitioner or leading a growing team, this talk provides a blueprint for using intelligence to build a more informed, resilient, and executive-aligned security organization.
Strip away the crayon drawings and here is what K-12 in the U.S. is: an enterprise network serving tens of millions of users, holding Social Security numbers, medical records, disability status, and behavioral data on children, staffed by people whose primary job is not security, under near-constant attack. One breach in 2024 exposed the records of 60 million students and 10 million teachers in a single hit. The sector gets compromised at a rate most industries would declare a national emergency. Nobody is talking about it.
This talk draws on the experience of defending a large urban district serving over 92,000 students, 14,000 staff, and 138 schools. It makes the case that K-12 is not a niche problem. It is the most honest stress test available for defending a high-value, data-rich environment with enterprise-scale exposure and a fraction of enterprise-level resources. The threat patterns, governance failures, defender strategies, and hard lessons from K-12 translate directly to any organization doing more with less.
Who's running your team right now — you, or the tools you built to help you? Your AI doesn't panic at 2 am. It doesn't burn out, check out, or fail to have the hard conversation. But it also can't lead. And right now, neither can a lot of the humans beside it.
Look around your team. Understaffed. Overextended. Burning through people faster than you can replace them. New talent who've never had a real mentor, never learned to lead under pressure, never had the kind of hard conversation that actually builds judgment — because nobody had the time or the skills to teach them. This isn't bad luck. It's the accumulated damage of a pandemic that vaporized two years of in-person mentorship, followed by an AI wave that automated enough of the technical work that we convinced ourselves the human layer underneath didn't need investment anymore. We were wrong.
The data is unambiguous: two-thirds of organizations report critical skills gaps, cybersecurity job satisfaction has dropped 8 points in two years, and hiring managers are now ranking curiosity and problem-solving above technical certifications. The skills disappearing fastest aren't on any exam — they're judgment under pressure, leading through ambiguity, holding a team together when the playbook runs out. We didn't just lose those skills. We stopped teaching them. In some cases, we selected against them.
This session is part diagnosis, part field manual for practitioners operating in the real landscape — not the conference brochure version of it. We'll cover exactly what broke, why it's worse in security than most fields, and how to use AI — the very thing accelerating the atrophy — to rebuild the human skills no algorithm can replace. Because at 2 am during a crisis, when the alerts won't stop, and someone in the room is about to break — something in that room has to be human enough to hold it together. That part doesn't get automated. It gets developed. Or it gets lost.
Break
Break
Break
SO-CRATES is a new free and open source project for analyzing pcap files, log files, and binary files. Designed from first principles, it's a single container image that is simple yet tremendously powerful. In this talk, we'll discuss the reasons for building a new tool, how it compares to the full Security Onion platform, and then see how you can view network alerts, file alerts, and log alerts. We'll then dive deeper into browsing network metadata (DNS, HTTP, TLS, flows), extracting ASCII transcripts, viewing per-packet hexdumps, and carving individual streams.
You deployed an AI service, followed the docs, and even remembered to rotate your passwords (well - mostly). But somewhere on the internet, a criminal syndicate is using your Azure API key to generate content you’d rather not explain to HR - and they’re billing it to you. This session walks through the Storm-2139 takedown - a real Microsoft Digital Crimes Unit operation spanning four countries, a federal lawsuit, and criminals who responded by harassing Microsoft’s lawyers online (yes, really). We’ll cover how AI APIs became the hottest new attack surface, and what you can actually do about it before the bill arrives.
AI agents don't attack like software, and they don't fail like humans. After years spent breaking traditional systems and now agentic ones I've watched the industry try to bolt old defenses onto a fundamentally new attack surface. It doesn't work, and the attack chains prove it.
This talk walks through the attacks actually hitting production AI agents today: privilege escalation through tool chaining, cross-session credential bleed, autonomy drift, and the ways a single compromised agent can pivot faster and further than any human attacker. We'll show why these don't map onto existing frameworks and why treating an agent like a user, or like an API, gets you owned either way.
From there we build the AI Agent Kill Chain: a practical model for where these attacks can actually be interrupted, and what stops them for real architecture, not policy documents. You'll leave with a holistic strategy you can apply Monday morning, not another slide of best practices.
Along the way, we'll call out what the AI security market is currently selling versus what it's actually stopping because right now most of it is marketing wrapped around old tooling, and that gap is exactly where the breaches are happening.
Security workflows are shifting from human-centric to AI-centric. Meanwhile, AI coding assistants have democratized software development, transforming domain expertise and creative ideas into powerful solutions without requiring a CS degree, and without consequences if things don’t work out.
What happens when these trends collide?
Experimentation with AI-enabled security tools and workflows becomes practically mandatory. The risk is minimal; the potential rewards are great.
Members of Cisco’s Talos team will share case studies from their research in building AI-driven security workflows. Discover frameworks for integrating AI into your operations and learn strategies to guide your customers through their AI transformations.
While reviewing my own promotion records in a Department of Defense web application, I noticed sensitive identifiers being passed in backend requests. Acting as a fully authenticated but non-privileged user, this curiosity revealed a widespread access control failure within an internal personnel system.
This talk examines how common vulnerabilities—such as indirect object reference flaws and over-trusted authenticated users—can expose massive amounts of Personally Identifiable Information without exploits, malware, or elevated access. Limited, ethical testing demonstrated access paths to sensitive records across multiple DoD populations. All findings were responsibly disclosed through official vulnerability channels.
Rather than focusing on specific technologies, the presentation highlights repeatable failure patterns in internal applications, why “trusted user” assumptions break down, and how curiosity-driven analysis uncovers issues that compliance testing often misses. Attendees will leave with a stronger mindset for evaluating internal systems, improving defensive testing programs, and understanding why internal vulnerability discovery is critical to modern security.
The expected path is where a simulated environment feels most convincing. The real test begins when someone treats it like an open world, wanders off the route, and pushes toward what should be the edge. This talk asks whether the illusion can hold when the machine being explored does not actually exist: no provisioned system, no command execution, just believable responses produced in real time. For deception, training, and security research, that kind of sustained exploration is valuable, but traditional labs make it expensive because every believable layer has to be built and maintained, especially for legacy hosts, industrial systems, carrier-grade equipment, or exotic architectures. This talk explores a just-in-time approach that uses guided language models to render the world around the user instead of building it in advance. Through real examples and a demo of the working implementation, I will share design choices, failure modes, and lessons from a system that must generate, remember, and react in real time. The result is a different way to think about simulated environments: not building every possible branch in advance, but creating believable depth at the moment exploration demands it.