TRAINING COURSES
Adversary Intelligence: LLM Tradecraft
WHAT TO EXPECT
Close the gap between using AI and understanding it
AI capability development and adoption are outpacing security teams’ ability to evaluate and secure these systems. Practitioners who understand how LLMs work, where they fail, and how adversaries use them are better positioned to apply AI effectively and defend the systems built around it. This course builds that technical foundation through LLM fundamentals, agent architecture, evaluation, threat modeling, offensive LLM tradecraft, and AI-assisted reverse engineering.
key takeaways
Course summary
Agentic systems are already in the environments you’re responsible for securing. Understanding how they work, where they fail, and how they can be exploited is no longer optional.
In this course, you’ll learn how modern LLMs process input and generate output, how prompting strategies affect model behavior, and why the architectural properties of these systems create security implications that don’t map cleanly onto traditional network security thinking. From there, the focus shifts to agentic systems: how they orchestrate tools and memory, where those patterns create exploitable gaps, and how attackers abuse them through prompt injection, jailbreaks, and insecure AI infrastructure. Labs run throughout every module, giving you hands-on experience with leading AI tooling across LLM system security, threat modeling, reverse engineering, and defensive security workflows.
Participants will learn
How to build and evaluate LLMs and agentic systems, including their architecture, limitations, and security implications in real enterprise environments
How to identify and understand attacker techniques targeting LLMs and agentic systems, including prompt injection, jailbreaks, and exploitable weaknesses in AI infrastructure and agent design
How to apply AI to defensive security workflows, including threat modeling, security testing, and AI-assisted reverse engineering using leading AI technologies
A CLOSER LOOK AT THE COURSE
Adversary Intelligence: LLM Tradecraft
The gap between using AI and understanding it is where most security risk lives. This course is built to close that.
This course gives security practitioners the technical depth to work confidently with modern LLMs and agentic systems and understand how adversaries approach them. The course opens with how large language models work: tokenization, context windows, attention mechanisms, and the architectural properties that create security implications traditional network security thinking doesn’t account for. From there, it builds into prompting strategies, agent architecture, and the infrastructure agentic systems rely on, including tool calling, memory, multi-agent coordination, and MCP servers.
Dig into LLM Tradecraft
An attacker perspective is woven throughout. Participants learn how adversaries are using and exploiting these systems through prompt injection, jailbreaks, and weaknesses in AI infrastructure including insecure MCP configurations and overprivileged agent identities. Defensive patterns are addressed at the input, output, and infrastructure levels. Observability and evaluation methods are integrated across the course, giving participants tools to assess model behavior and measure the reliability of systems they build or inherit.
Labs run throughout every module using Codex and other leading AI technologies, covering agentic system development, threat modeling, reverse engineering and defensive security workflows. Participants leave with the technical depth to evaluate AI systems critically, hands-on experience applying leading AI technologies to real security workflows, and practical skills for assessing and securing agentic systems in real enterprise environments.
Here’s what we’ll cover:
- AI foundations and the evolution of machine learning, deep learning, and LLMs
- Tokenization, embeddings, attention mechanisms, context windows, and architectural limitations
- Prompting strategies and prompt-as-program design
- Agent architecture, tool calling, memory, MCP servers, and multi-agent coordination
- Coding agents and Codex workflows
Here’s what we’ll cover:
- LLM observability and evaluation
- Threat modeling for LLM systems
- Jailbreaks, prompt injection, and defensive patterns
- Attacking and securing AI infrastructure
- AI-assisted reverse engineering workflows
Overview
Dig into LLM Tradecraft
An attacker perspective is woven throughout. Participants learn how adversaries are using and exploiting these systems through prompt injection, jailbreaks, and weaknesses in AI infrastructure including insecure MCP configurations and overprivileged agent identities. Defensive patterns are addressed at the input, output, and infrastructure levels. Observability and evaluation methods are integrated across the course, giving participants tools to assess model behavior and measure the reliability of systems they build or inherit.
Labs run throughout every module using Codex and other leading AI technologies, covering agentic system development, threat modeling, reverse engineering and defensive security workflows. Participants leave with the technical depth to evaluate AI systems critically, hands-on experience applying leading AI technologies to real security workflows, and practical skills for assessing and securing agentic systems in real enterprise environments.
Foundations
Here’s what we’ll cover:
- AI foundations and the evolution of machine learning, deep learning, and LLMs
- Tokenization, embeddings, attention mechanisms, context windows, and architectural limitations
- Prompting strategies and prompt-as-program design
- Agent architecture, tool calling, memory, MCP servers, and multi-agent coordination
- Coding agents and Codex workflows
Security Applications
Here’s what we’ll cover:
- LLM observability and evaluation
- Threat modeling for LLM systems
- Jailbreaks, prompt injection, and defensive patterns
- Attacking and securing AI infrastructure
- AI-assisted reverse engineering workflows
Before you attend
Who should attend
This course is intended for security practitioners, researchers, engineers, defenders, and technical leaders who need to understand how modern LLMs and agentic systems work, how adversaries target them, and how to evaluate and secure AI-enabled workflows in enterprise environments.
Prerequisites
Participants should be comfortable with technical security concepts and basic command-line workflows. Prior LLM security experience is not required, though familiarity with programming, security testing, threat modeling, or reverse engineering will help participants get the most from the labs.
What to bring
Participants must provide their own computer with a modern web browser installed to access training materials and complete the course’s labs. The SpecterOps training platform domain (specterops.training) must be accessible from the participant’s computer throughout the duration of the course.
There are no local virtual machines required, but participants should be prepared to use browser-based lab environments, command-line tooling, and guided AI-assisted workflows during hands-on exercises.
What you receive
During the course, participants receive access to the training platform, including all pre-recorded lecture content, hands-on lab environments, guided exercises, and supporting materials.
Upon completion of the course, participants receive:
- A copy of the course slides
- A certificate of completion
- A digital badge
Accepting your digital badge confirms your SpecterOps Training alumni status, which conveys exclusive discounts to future SpecterOps hosted training.
MORE WAYS TO TRAIN
Private and custom training
SpecterOps courses, delivered exclusively for your team. Need something beyond our current offerings? We develop custom curriculum, labs, and CTFs designed around your team’s specific goals and threat landscape. Our training is taught by the same front-line practitioners who conduct our engagements, bringing real-world experience into every course.
DEEPEN YOUR TRADECRAFT
Explore additional training courses
Adversary Perspectives: Active Directory
Learn Active Directory’s architecture and security implications, and identify misconfigurations before an attacker does.
Adversary Perspectives: Azure
Discover how adversaries view, target, and exploit Azure and Entra ID environments.
Adversary Tactics: Red Team Operations
Go beyond Domain Admin and sharpen your offense-in-depth skills.
Adversary Tactics: Identity-Driven Offensive Tradecraft
What turns a path into an attack path? Learn how to find and abuse them.
Adversary Tactics: Detection
Stop chasing indicators. Build detections that focus on how attackers operate.
Adversary Tactics: Tradecraft Analysis
Deconstruct how attack techniques really work, then build detections or learn how to evade them.
SpecterOps Tradecraft Academy
Hands-on offensive and defensive security training built by SpecterOps practitioners, available on demand and designed to be completed at your own pace.
