Yeo JooHo
Lead Researcher | PIOLINK Cybersecurity Research Team
Seoul, South Korea
Actions
YEO JOOHO is a seasoned Security Researcher and White Hat Hacker with a distinguished career spanning military operations, government investigations, and large-scale cloud infrastructure security. He has served as a cyber threat analyst for major CSPs (Cloud Service Providers) and provided critical technical support for high-profile anti-corruption investigations at the national prosecutor's office.
A graduate of the prestigious Best of the Best (BoB) next-generation security leader program, his offensive and defensive capabilities are globally recognized. His accolades include a 2nd place finish (Blue Team) in the National Cyber Conflict Exercise (CCE) and multiple commendations for excellence in cloud incident response.
As an active vulnerability researcher, Jooho has a prolific history of discovering and reporting security flaws in open-source web frameworks, mobile messengers, and enterprise systems. He has also contributed to national security as an official challenge author for military cyber warfare competitions and an advisory member for government-led research projects.
Area of Expertise
Topics
Breach on Autopilot: From AI-Planned Kill-Chains to Automated Atomic Execution
Modern adversaries execute multi-stage kill-chains, yet offensive tools remain fragmented. The disconnect between discovery and execution forces manual intervention, slowing Red Team operations. This session introduces a unified "Blue-to-Red" framework bridging asset intelligence (Blue Hunter) and automated execution (Red Hunter), transforming AI into a strategic operative.
Technical Docs: https://keen-harmonica-78a.notion.site/Breach-on-Autopilot-Technical-Deep-Dive-352d19e62e02802f99b8d72376af1bc8
Blue Hunter is the cognitive engine. It ingests OSINT—DNS, certificates, and tech stacks. It reasons through environmental contexts to calculate lateral movement probabilities. AI intent is algorithmically mapped to MITRE ATT&CK TIDs, grounding plans in proven tradecraft.
Red Hunter handles action. Its Dynamic Parser ingests Atomic Red Team YAMLs, injecting target parameters in real-time. Using an RPC bridge to Metasploit, it enables programmatic session upgrades and post-exploitation, replacing manual console work with a seamless loop.
The framework operates in air-gapped environments by serializing MITRE ATT&CK and Atomic repositories into a local vector database. Using local LLMs, it ensures high-fidelity operations without external API dependencies.
The pipeline generates a MITRE ATT&CK Coverage Matrix to close "Detection Gaps." Automating the OODA loop reduces manual integration from hours to minutes, shifting the focus to data-driven strategy.
Breach on Autopilot: From AI-Planned Kill-Chains to Automated Atomic Execution
A full-stack offensive framework that bridges AI-driven attack surface intelligence with automated, air-gapped red team execution — end-to-end, with minimal human
intervention.
Breach on Autopilot: From AI-Planned Kill-Chains to Automated Atomic Execution
Session Description
[Please refer to this link for detailed technical documentation, architecture diagrams: [https://keen-harmonica-78a.notion.site/Breach-on-Autopilot-Technical-Deep-Dive-352d19e62e02802f99b8d72376af1bc8?source=copy_link]
General Information
- Title: Breach on Autopilot: From AI-Planned Kill-Chains to Automated Atomic Execution
- Category: Artificial Intelligence (AI) Hacking / Enterprise Security
- Keywords: Autonomous Red Teaming, LLM Reasoning, MITRE ATT&CK, Atomic Red Team, Air-gapped Security, Offensive AI
- Project Documentation: [Breach on Autopilot: Technical Deep Dive](https://www.notion.so/Breach-on-Autopilot-Technical-Deep-Dive-352d19e62e02802f99b8d72376af1bc8?pvs=21) (click)
- Moving Beyond the AI Buzzword: Building an Autonomous Offensive Pipeline
Modern attackers execute coordinated, multi-stage kill-chains, yet many offensive tools remain fragmented. This session introduces an in-house automated offensive framework that bridges the gap between asset intelligence (Blue Hunter) and autonomous execution (Red Hunter).
We will demonstrate a unified "Blue-to-Red" workflow where AI doesn't just generate text, but acts as a strategic operative. We introduce the AI Dependency Logic, a tiered model allowing users to scale from manual validation to full autonomy. In "High Autonomy" mode, the framework performs Auto-Resolution—analyzing execution failures to self-correct payloads—and Intelligent Pathfinding to navigate restricted environments via automated credential injection and alternative maneuvers.
Key Technical Highlights:
- Strategic Pivot: A seamless interface to transition from Blue Hunter’s adversary mapping and CVE prioritization to Red Hunter’s tactical execution.
- Stealth Agent Generation: AI-driven defense evasion that dynamically applies obfuscation and anti-analysis measures based on the target's EDR posture.
- Air-Gapped Resilience: A fully offline architecture utilizing local LLMs and serialized MITRE ATT&CK/Atomic Red Team caches for high-security environments.
- Tactical Analytics: Transforming raw execution logs into a localized MITRE ATT&CK Coverage Matrix to identify explicit detection gaps.
Join us to see how AI moves beyond the buzzwords into a self-healing, reproducible offensive pipeline that helps IT teams reclaim their time and focus on strategic defense.
Speakers Introduction
Jooho Yeo (Researcher, PIOLINK) & Yeonju Baek (Developer & Researcher, PIOLINK)
We are a team from PIOLINK, a cybersecurity company based in South Korea.
First, let me introduce Jooho Yeo. Jooho is a researcher specializing in AI-driven Attack Surface Management. He is the mastermind behind Blue Hunter, where he focuses on transforming raw OSINT data into strategic attack narratives using LLMs. He’s the one who teaches our AI how to 'think' like a strategist.
And I am Yeonju Baek. I am a developer and researcher focused on offensive automation. My role is leading the development of Red Hunter, the execution engine of this framework. My work is about operationalizing the MITRE ATT&CK framework and Atomic Red Team payloads into seamless pipelines that work even in the most restricted, air-gapped environments.
Together, we’ve built a bridge between AI-generated intelligence and real-world offensive execution.
---
PART 1: Blue Hunter – AI Strategic Reasoning
Meet Blue Hunter, our AI-driven strategic brain. It ingests OSINT data—DNS, certificates, tech-stacks—and processes it through a Chain-of-Thought prompting strategy. Instead of just listing vulnerabilities, it reasons: 'If this service is EOL and this port is open, what is the probability of a successful lateral movement?' It generates an actual 'Operation Log' that describes a full, environment-specific narrative.
This narrative isn't just text. Our engine algorithmically maps these identified weaknesses to specific MITRE ATT&CK TIDs. It bridges the gap between high-level AI intent and low-level system commands. This ensures that every move the AI plans is grounded in proven threat actor tradecraft.
---
PART 2: Red Hunter – Automated Execution
Once the plan is ready, Red Hunter takes over. It’s an automated execution engine that supports both agentless and agent-based operations. It features a Dynamic Parser that takes Atomic Red Team YAMLs and injects target-specific parameters like IPs and credentials. It doesn't just run a script; it orchestrates an operation.
Execution in complex networks requires persistence. Red Hunter uses an RPC bridge to Metasploit, allowing us to handle session upgrades and post-exploitation modules programmatically. We’re moving away from manual console typing to a fully automated session handling loop.
---
PART 3: Engineering for Isolation
Here is the core technical challenge: How do you do this in a zero-trust, air-gapped environment? We built a localized intelligence system. We serialized the entire MITRE tree and Atomic repositories into a local cache. No GitHub, no cloud APIs. The AI searches a pre-computed vector database stored right on the local disk.
---
PART 4: Impact & Metrics
To bridge the gap with the Blue Team, we generate a Coverage Matrix. This visualizes exactly which TIDs passed, failed, or were partially blocked. It highlights the 'Detection Gaps' in the target's environment. We’re not just providing a list of bugs; we’re providing a roadmap for better defense.
In conclusion, Blue Hunter plans, and Red Hunter executes. By automating the OODA loop, we reduced manual data integration from hours to minutes. Our next step is a fully self-evolving framework. Security is no longer a battle of tools; it’s a battle of data-driven strategy.
Breach on Autopilot: Closing the CTEM Validation Gap with Autonomous Adversary Emulation
General Information
Category: Artificial Intelligence (AI) in Security / Offensive Cyber Security Operations
Keywords: Autonomous Adversary Emulation, LLM Reasoning, MITRE ATT&CK, Air-Gapped Security, Security Control Validation, Purple Teaming
Moving Beyond the AI Buzzword: Building an Autonomous Offensive Pipeline
Modern threat actors execute coordinated, context-aware, multi-stage kill-chains, yet many automated offensive tools remain fragmented and rely on brittle, static scripts. This session introduces an in-house, autonomous offensive framework designed to bridge the gap between actual asset intelligence (Blue Hunter) and stateful, automated execution (Red Hunter) without relying on external cloud APIs.
We will demonstrate a unified "Blue-to-Red" validation workflow where AI does not merely generate text, but acts as a deterministic strategic operative. We introduce the Dual Autonomy Architecture, a tiered model allowing operators to scale from strictly controlled deterministic validation to full adversary emulation. In "High Autonomy" mode, the framework utilizes an error-based State Machine to perform Auto-Resolution—capturing execution failures (e.g., EDR blocks or missing dependencies) and feeding them back into local LLMs to dynamically mutate payloads and pivot attack chains in real-time.
-----------------------------------------------------------------------------------------
Key Technical Highlights:
Context-Aware Strategic Pivot: A seamless programmatic bridge that transitions from the cognitive engine’s ASM (Attack Surface Management) data and CVE prioritization directly into tactical execution, automatically classifying and routing techniques into Agent or Agentless attack paths based on target reachability.
Target-Dependent Payload Pre-rendering: Eliminating runtime mapping errors by having the AI intelligence engine dynamically pre-render "ready-to-fire" payloads with target-specific variables (IPs, open ports, credentials) completely hardcoded prior to execution.
Air-Gapped Resilience (Zero Data Leak): A fully offline architecture utilizing quantized local LLMs and serialized MITRE ATT&CK/Atomic Red Team vector caches, designed for highly regulated environments that strictly prohibit external data transmission.
Tactical Analytics & Detection Engineering: Transforming raw execution logs and self-healing mutation attempts into a localized, empirical MITRE ATT&CK Coverage Matrix, explicitly pinpointing true detection gaps for Purple Teams.
Join us to explore the engineering depths of how AI moves beyond theoretical concepts into a reproducible, self-healing offensive pipeline that empowers defenders to empirically validate their security posture and focus on threat-informed defense.
-----------------------------------------------------------------------------------------
From Asset Cognition to Real-Time Self-Healing: Designing and Implementing a Fully Offline Autonomous Offensive Framework to Close the CTEM Validation Gap
As Continuous Threat Exposure Management (CTEM) establishes itself as the standard for modern enterprise security, asset discovery technologies that identify attack surfaces have become highly automated. However, the "Validation" phase—proving whether an identified vulnerability is actually exploitable in a target environment—remains a severe bottleneck, heavily reliant on the manual intervention and resources of defense teams.
To automate this validation process, most Breach and Attack Simulation (BAS) tools widely used in the industry today remain stuck in a binary approach, blindly projecting pre-compiled, static atomic scripts against targets. This static approach causes fatal crashes in the face of minor environmental changes, such as slight patch version differences in the target OS, missing dependency libraries, or the intervention of modern Endpoint Detection and Response (EDR) solutions. This results in simulation failures caused not by superior security controls, but by the "brittleness" of the execution scripts themselves. Conversely, while there are attempts to introduce external cloud-based Large Language Model (LLM) APIs into the offensive pipeline to overcome this lack of flexibility, this requires transmitting sensitive internal network topologies and asset vulnerability data to external servers. Consequently, its adoption is fundamentally impossible in financial and public infrastructures that strictly mandate Zero Trust and Zero Data Leak architectures.
In this session, we will unveil for the first time the core architecture and engineering mechanics of a fully offline-based autonomous offensive framework that perfectly overcomes both the static limitations of existing BAS systems and the data leak risks of cloud AI. This framework consists of the programmatic integration of a cognitive engine (Blue Hunter), which formulates strategies based on actual asset data exposed on the internet rather than a virtual sandbox, and an execution engine (Red Hunter), which controls the execution flow by autonomously analyzing errors. Specifically, to perform top-tier reasoning and code generation even in environments completely disconnected from external networks, the system operates code-generation-specific models and custom-finetuned cybersecurity models locally on its own Ollama server. Through this, it satisfies the absolute Zero Data Leak requirement, ensuring that not a single byte of customer data is leaked externally.
The first core of the framework is its multi-source-based Attack Surface Management (ASM) pipeline and rule-based asset normalization. Moving beyond simple port scanning, the cognitive engine resolves DNS records based on the root domain and correlates internet-exposed ports and Common Vulnerabilities and Exposures (CVE) data. Simultaneously, it collects vast amounts of unstructured OSINT data, including subdomain extraction via Certificate Transparency logs and historical URL patterns. The collected heterogeneous data goes through a rule evaluation engine that performs multi-layer matching of banner information, HTTP response headers, TLS subject strings, and CPE identifiers using regular expressions, rather than simple version matching, to meticulously construct the target's risk profile.
This established real-world asset context then passes through an intelligent payload classification algorithm. To overcome the limitations of existing tools that mandatorily require agents to be installed on target systems or stop at external network scanning, the framework operates a hybrid heuristic algorithm combining a Tactic and Technique ID (TID) dictionary and behavioral keywords within the AI-generated attack scenarios. Through this, the framework autonomously determines and dynamically routes whether the technique requires an Agent method executing inside the target, or an Agentless method capable of striking from the outside. Furthermore, to eradicate variable mapping errors at runtime, the cognitive engine does not simply pass the TID to the execution engine. For each attack technique object, it directly generates a "Ready-to-fire," target-dependent payload with the asset context—such as the actual IP, identified FQDN, and open ports—completely hardcoded, encapsulating it in a JSON schema.
The formulated scenario is handed over to the execution engine via a signature-based one-way Single Sign-On (SSO) architecture that bridges the two engines. Protected by timestamps and Hash-based Message Authentication Codes (HMAC), this bridge allows security analysts to instantly transition from the asset analysis dashboard to the red teaming operational environment without additional authentication or context switching.
Upon receiving the scenario, the execution engine automatically triggers a campaign based on the identified actual CVE list. Using a mapping dictionary built on NVD API data, CVEs are translated into Common Weakness Enumerations (CWEs), which are automatically mapped to specific MITRE ATT&CK TID sequences corresponding to OS Command Injection, SQL Injection, Authentication Bypass, and more. The execution engine's most powerful engineering achievement lies in its Dual Autonomy mechanism, which operators can select based on the operational environment, and its State Machine based on standard error (stderr) parsing.
In environments requiring extreme predictability, the Low Autonomy mode is engaged. Upon execution failure, it suppresses AI generation and safely falls back to numerous static atomic payloads pre-loaded in the framework. Conversely, in High Autonomy mode—which emulates the flexibility of a real attacker—a true autonomous execution loop unfolds. If execution is blocked by a security solution after payload delivery, the execution engine does not simply log a failure and halt. It parses the returned error log in real-time and feeds it back to the local AI engine.
Finally, to fundamentally prevent destructive outages on production networks that complete autonomy might cause, a rigorous Risk Assessor module is implemented within the execution layer. All payloads dynamically mutated by the local AI must pass through this static heuristic engine immediately before being transmitted to the target. It evaluates risk weights by scanning for the presence of fatal system destruction keywords, such as disk formatting, deletion of critical system files, or complete deactivation of network firewalls. If the calculated risk score exceeds a configured threshold, the framework immediately suspends execution at the API level and activates a Human-in-the-Loop control gate that mandates approval from a human operator.
Through this session, attendees will discover the technical depths of how fragmented ASM data translates into deterministic execution payloads, and how disconnected attack scripts evolve into a living, organic adversary emulator when merged with a state machine and local artificial intelligence.
Breach on Autopilot: Building an Offline Autonomous Offensive Pipeline for CTEM Validation
As Continuous Threat Exposure Management (CTEM) becomes a staple in enterprise security, the "Validation" phase—proving whether an exposure is practically exploitable—remains a manual bottleneck. Existing Breach and Attack Simulation (BAS) tools often rely on brittle, static scripts that crash against minor environmental differences or modern EDRs. Conversely, integrating cloud-based AI to automate this process introduces severe data privacy risks, making it unsuitable for air-gapped or strictly regulated environments.
This session shares our engineering journey of building an in-house, fully offline autonomous offensive framework designed to close the validation gap safely. We will explore how to bridge a cognitive engine (ASM and intelligence) with an execution engine (stateful exploitation) using local LLMs (e.g., via Ollama). Attendees will learn how we parse unstructured OSINT data to pre-render target-dependent payloads, and how we engineered a bounded State Machine that parses stderr logs in real-time. Instead of abruptly crashing, the framework performs "Intra-technique Self-healing"—dynamically mutating the payload's syntax to bypass blocks without hallucinating out-of-scope attacks.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top