diff --git a/docs/AI_LLM Red Team Field Manual.md b/docs/AI_LLM Red Team Field Manual.md index 08d4099..c3feea0 100644 --- a/docs/AI_LLM Red Team Field Manual.md +++ b/docs/AI_LLM Red Team Field Manual.md @@ -1,6 +1,6 @@ # **AI/LLM Red Team Field Manual** -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) > **For Junior Penetration Testers**: This manual is designed as a complete, standalone field guide. Follow the Quick Start below to begin testing within 15 minutes. @@ -188,7 +188,7 @@ For step-by-step attack procedures with extensive code examples, see the **modul --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **1. Introduction: Rules of Engagement (RoE)** @@ -206,7 +206,7 @@ Define in writing: in-scope systems/models, allowed techniques, test windows, ha --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **1.5 Environment Setup & Configuration** @@ -493,7 +493,7 @@ API Config File: ✅ --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **2. Red Teaming Phases** @@ -772,7 +772,7 @@ chmod +x cleanup.sh --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **3. Attack Types & Practical Test Examples** @@ -1046,7 +1046,7 @@ radamsa input_sample.json | curl -d @- $API_URL --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **4\. Tools Reference & CLI Commands** @@ -1066,7 +1066,7 @@ radamsa input_sample.json | curl -d @- $API_URL --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **5. Attack-Type–to–Tool Quick Reference** @@ -1089,7 +1089,7 @@ See [Quick Reference Card](field_manuals/Field_Manual_Quick_Reference.md) for th --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **5.5 API Configuration Guide** @@ -1357,7 +1357,7 @@ echo "✅ Provider test complete!" --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **6\. Reporting Guidance** @@ -1371,7 +1371,7 @@ Report every finding with: --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **7\. Additional Guidance** @@ -1383,7 +1383,7 @@ Report every finding with: --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **8. Troubleshooting Guide** diff --git a/docs/Chapter_01_Introduction_to_AI_Red_Teaming.md b/docs/Chapter_01_Introduction_to_AI_Red_Teaming.md index 56326be..8528f89 100644 --- a/docs/Chapter_01_Introduction_to_AI_Red_Teaming.md +++ b/docs/Chapter_01_Introduction_to_AI_Red_Teaming.md @@ -12,7 +12,7 @@ Related: Chapters 2 (Ethics), 3 (Mindset), 7 (Lab Setup) # Chapter 1: Introduction to AI Red Teaming

- +

_This chapter introduces AI red teaming as a structured practice for identifying vulnerabilities in AI systems, particularly Large Language Models. You'll learn what makes AI security different from traditional cybersecurity, understand the unique attack surfaces of LLMs, and explore the professional mindset and ethical framework required for responsible AI security testing._ diff --git a/docs/Chapter_02_Ethics_Legal_and_Stakeholder_Communication.md b/docs/Chapter_02_Ethics_Legal_and_Stakeholder_Communication.md index 07ea0cb..fda3c59 100644 --- a/docs/Chapter_02_Ethics_Legal_and_Stakeholder_Communication.md +++ b/docs/Chapter_02_Ethics_Legal_and_Stakeholder_Communication.md @@ -12,7 +12,7 @@ Related: Chapters 4 (SOW/RoE), 8 (Evidence), 36 (Reporting) # Chapter 2: Ethics, Legal, and Stakeholder Communication

- +

_This chapter establishes the critical ethical and legal framework for AI red teaming. You'll learn the principles of responsible security testing, understand legal boundaries and authorization requirements, master stakeholder communication strategies, and develop professional practices for data handling and responsible disclosure in AI security engagements._ diff --git a/docs/Chapter_03_The_Red_Teamers_Mindset.md b/docs/Chapter_03_The_Red_Teamers_Mindset.md index 8bf0c0a..6d20801 100644 --- a/docs/Chapter_03_The_Red_Teamers_Mindset.md +++ b/docs/Chapter_03_The_Red_Teamers_Mindset.md @@ -12,7 +12,7 @@ Related: Chapters 2 (Ethics), 5 (Threat Modeling) # Chapter 3: The Red Teamer's Mindset

- +

_This chapter develops the critical thinking skills and adversarial perspective essential for effective AI red teaming. You'll learn how to think like an attacker, cultivate skepticism and creativity, balance persistence with professionalism, and develop the psychological resilience needed for this specialized security discipline._ diff --git a/docs/Chapter_04_SOW_Rules_of_Engagement_and_Client_Onboarding.md b/docs/Chapter_04_SOW_Rules_of_Engagement_and_Client_Onboarding.md index 8eb3447..af50c77 100644 --- a/docs/Chapter_04_SOW_Rules_of_Engagement_and_Client_Onboarding.md +++ b/docs/Chapter_04_SOW_Rules_of_Engagement_and_Client_Onboarding.md @@ -12,7 +12,7 @@ Related: Chapters 2 (Ethics), 6 (Scoping), 8 (Evidence) # Chapter 4: SOW, Rules of Engagement, and Client Onboarding

- +

_This chapter provides practical frameworks for establishing professional AI red team engagements. You'll learn to create comprehensive Statements of Work, define clear Rules of Engagement, conduct effective client onboarding, and establish communication protocols that ensure legal protection and professional success._ diff --git a/docs/Chapter_05_Threat_Modeling_and_Risk_Analysis.md b/docs/Chapter_05_Threat_Modeling_and_Risk_Analysis.md index d2dca3c..bd696c1 100644 --- a/docs/Chapter_05_Threat_Modeling_and_Risk_Analysis.md +++ b/docs/Chapter_05_Threat_Modeling_and_Risk_Analysis.md @@ -12,7 +12,7 @@ Related: Chapters 6 (Scoping), 9 (Architectures), 31 (Reconnaissance) # Chapter 5: Threat Modeling and Risk Analysis

- +

_This chapter teaches systematic threat modeling and risk analysis for AI systems. You'll learn to identify critical assets, enumerate threat actors, map attack surfaces specific to LLMs, prioritize risks by impact and likelihood, and create actionable threat models that drive effective red team engagements._ diff --git a/docs/Chapter_06_Scoping_an_Engagement.md b/docs/Chapter_06_Scoping_an_Engagement.md index 6f2d8cc..206c6c4 100644 --- a/docs/Chapter_06_Scoping_an_Engagement.md +++ b/docs/Chapter_06_Scoping_an_Engagement.md @@ -12,7 +12,7 @@ Related: Chapters 4 (SOW/RoE), 5 (Threat Modeling), 7 (Lab Setup) # Chapter 6: Scoping an Engagement

- +

_This chapter guides you through the critical process of scoping AI red team engagements. You'll learn to define realistic boundaries, estimate timelines and resources, identify technical and business constraints, communicate scope effectively to stakeholders, and avoid common scoping pitfalls that derail engagements._ diff --git a/docs/Chapter_07_Lab_Setup_and_Environmental_Safety.md b/docs/Chapter_07_Lab_Setup_and_Environmental_Safety.md index 278582b..0576573 100644 --- a/docs/Chapter_07_Lab_Setup_and_Environmental_Safety.md +++ b/docs/Chapter_07_Lab_Setup_and_Environmental_Safety.md @@ -12,7 +12,7 @@ Related: Chapters 9 (Architectures), 32 (Automation), 33 (Red Team Frameworks) # Chapter 7: Lab Setup and Environmental Safety

- +

_This chapter provides hands-on guidance for setting up safe, isolated AI red teaming environments. You'll learn to configure local and cloud-based labs, implement proper network isolation, deploy test models and applications, establish monitoring and logging, and create reproducible test environments for ethical AI security research._ diff --git a/docs/Chapter_08_Evidence_Documentation_and_Chain_of_Custody.md b/docs/Chapter_08_Evidence_Documentation_and_Chain_of_Custody.md index 1cfc236..0572189 100644 --- a/docs/Chapter_08_Evidence_Documentation_and_Chain_of_Custody.md +++ b/docs/Chapter_08_Evidence_Documentation_and_Chain_of_Custody.md @@ -12,7 +12,7 @@ Related: Chapters 2 (Ethics), 36 (Reporting), 40 (Compliance) # Chapter 8: Evidence, Documentation, and Chain of Custody

- +

_This chapter establishes professional standards for evidence collection and documentation in AI red team engagements. You'll learn forensically-sound capture methods, proper chain-of-custody procedures, secure evidence storage and handling, comprehensive documentation practices, and legal requirements for preserving findings that may be used in compliance or legal contexts._ diff --git a/docs/Chapter_09_LLM_Architectures_and_System_Components.md b/docs/Chapter_09_LLM_Architectures_and_System_Components.md index 68c5a10..525485a 100644 --- a/docs/Chapter_09_LLM_Architectures_and_System_Components.md +++ b/docs/Chapter_09_LLM_Architectures_and_System_Components.md @@ -12,7 +12,7 @@ Related: Chapters 10 (Tokenization), 11 (Plugins), 12 (RAG) # Chapter 9: LLM Architectures and System Components

- +

_This chapter provides deep technical understanding of LLM architectures critical for red teaming. You'll learn transformer internals, attention mechanisms, embedding spaces, system prompts and context windows, model deployment patterns, and how architectural choices create specific attack surfaces that red teamers can exploit._ diff --git a/docs/Chapter_10_Tokenization_Context_and_Generation.md b/docs/Chapter_10_Tokenization_Context_and_Generation.md index bebf310..5cff86f 100644 --- a/docs/Chapter_10_Tokenization_Context_and_Generation.md +++ b/docs/Chapter_10_Tokenization_Context_and_Generation.md @@ -12,7 +12,7 @@ Related: Chapters 9 (Architectures), 14 (Prompt Injection), 18 (Evasion) # Chapter 10: Tokenization, Context, and Generation

- +

_This chapter explores how LLMs process and generate text, with focus on security implications. You'll learn tokenization mechanisms (BPE, WordPiece), context window management, generation strategies (greedy, sampling, beam search), and how understanding these processes enables sophisticated attacks like token manipulation and evasion techniques._ diff --git a/docs/Chapter_11_Plugins_Extensions_and_External_APIs.md b/docs/Chapter_11_Plugins_Extensions_and_External_APIs.md index 5a53133..2a4eca4 100644 --- a/docs/Chapter_11_Plugins_Extensions_and_External_APIs.md +++ b/docs/Chapter_11_Plugins_Extensions_and_External_APIs.md @@ -12,7 +12,7 @@ Related: Chapters 12 (RAG), 17 (Plugin Exploitation), 26 (Supply Chain) # Chapter 11: Plugins, Extensions, and External APIs

- +

_This chapter examines the plugin and API ecosystem that extends LLM capabilities and creates new attack surfaces. You'll learn plugin architectures, function calling mechanisms, API integration patterns, authentication and authorization flows, and the unique vulnerabilities introduced when LLMs orchestrate external tool usage._ diff --git a/docs/Chapter_12_Retrieval_Augmented_Generation_RAG_Pipelines.md b/docs/Chapter_12_Retrieval_Augmented_Generation_RAG_Pipelines.md index 535a7a4..86dfd98 100644 --- a/docs/Chapter_12_Retrieval_Augmented_Generation_RAG_Pipelines.md +++ b/docs/Chapter_12_Retrieval_Augmented_Generation_RAG_Pipelines.md @@ -12,7 +12,7 @@ Related: Chapters 11 (Plugins), 13 (Supply Chain), 23 (Persistence) # Chapter 12: Retrieval-Augmented Generation (RAG) Pipelines

- +

_This chapter dissects Retrieval Augmented Generation systems and their attack surfaces. You'll learn RAG architecture (indexing, embedding, retrieval, generation), vector database security, context injection through retrieval poisoning, prompt leakage via retrieved documents, and how to test the complex data flow that makes RAG both powerful and vulnerable._ diff --git a/docs/Chapter_13_Data_Provenance_and_Supply_Chain_Security.md b/docs/Chapter_13_Data_Provenance_and_Supply_Chain_Security.md index 2d610eb..52f58b7 100644 --- a/docs/Chapter_13_Data_Provenance_and_Supply_Chain_Security.md +++ b/docs/Chapter_13_Data_Provenance_and_Supply_Chain_Security.md @@ -12,7 +12,7 @@ Related: Chapters 11 (Plugins), 19 (Training Data Poisoning), 26 (Supply Chain A # Chapter 13: Data Provenance and Supply Chain Security

- +

_This chapter addresses the critical but often overlooked aspect of AI supply chain security. You'll learn to trace data and model provenance, identify supply chain attack surfaces (datasets, pre-trained models, dependencies), assess third-party components, verify model integrity, and establish security controls that protect against poisoned training data and compromised model artifacts._ diff --git a/docs/Chapter_14_Prompt_Injection.md b/docs/Chapter_14_Prompt_Injection.md index ca5ed96..df58ee4 100644 --- a/docs/Chapter_14_Prompt_Injection.md +++ b/docs/Chapter_14_Prompt_Injection.md @@ -12,7 +12,7 @@ Related: Chapters 15-16 (Data Leakage, Jailbreaks) # Chapter 14: Prompt Injection (Direct/Indirect, 1st/3rd Party)

- +

_This chapter provides comprehensive coverage of prompt injection attacks, including direct and indirect injection techniques, first-party and third-party variations, detection methods, defense-in-depth strategies, real-world case studies, and critical ethical considerations for authorized security testing._ diff --git a/docs/Chapter_15_Data_Leakage_and_Extraction.md b/docs/Chapter_15_Data_Leakage_and_Extraction.md index d3cdef6..f4532b2 100644 --- a/docs/Chapter_15_Data_Leakage_and_Extraction.md +++ b/docs/Chapter_15_Data_Leakage_and_Extraction.md @@ -12,7 +12,7 @@ Related: Chapters 16, 20 (Jailbreaks, Model Theft) # Chapter 15: Data Leakage and Extraction

- +

_This chapter provides comprehensive coverage of data leakage vulnerabilities in LLM systems, including training data extraction, conversation history leakage, system prompt disclosure, credential extraction, PII revelation, model inversion attacks, detection strategies, mitigation techniques, and critical regulatory compliance considerations._ diff --git a/docs/Chapter_16_Jailbreaks_and_Bypass_Techniques.md b/docs/Chapter_16_Jailbreaks_and_Bypass_Techniques.md index ad1bb4b..0c19b7a 100644 --- a/docs/Chapter_16_Jailbreaks_and_Bypass_Techniques.md +++ b/docs/Chapter_16_Jailbreaks_and_Bypass_Techniques.md @@ -12,7 +12,7 @@ Related: Chapters 11, 17 (Safety/Alignment, Plugin Exploitation) # Chapter 16: Jailbreaks and Bypass Techniques

- +

_This chapter provides comprehensive coverage of jailbreak techniques, bypass methods, testing methodologies, and defenses for LLM systems._ diff --git a/docs/Chapter_17_01_Fundamentals_and_Architecture.md b/docs/Chapter_17_01_Fundamentals_and_Architecture.md index 6aacb41..9a4c513 100644 --- a/docs/Chapter_17_01_Fundamentals_and_Architecture.md +++ b/docs/Chapter_17_01_Fundamentals_and_Architecture.md @@ -12,7 +12,7 @@ Related: Chapter 15 (Data Leakage), Chapter 23 (Persistence) # Chapter 17: Plugin and API Exploitation

- +

_This chapter covers security issues in LLM plugins, APIs, and third-party integrations—from architecture analysis and vulnerability discovery to exploitation techniques and defense strategies._ diff --git a/docs/Chapter_18_Evasion_Obfuscation_and_Adversarial_Inputs.md b/docs/Chapter_18_Evasion_Obfuscation_and_Adversarial_Inputs.md index d62728b..0709762 100644 --- a/docs/Chapter_18_Evasion_Obfuscation_and_Adversarial_Inputs.md +++ b/docs/Chapter_18_Evasion_Obfuscation_and_Adversarial_Inputs.md @@ -12,7 +12,7 @@ Related: Chapter 16 (Jailbreaks), Chapter 21 (DoS) # Chapter 18: Evasion, Obfuscation, and Adversarial Inputs

- +

_This chapter provides comprehensive coverage of evasion techniques, obfuscation methods, and adversarial input strategies used to bypass LLM security controls, along with detection and mitigation approaches._ diff --git a/docs/Chapter_19_Training_Data_Poisoning.md b/docs/Chapter_19_Training_Data_Poisoning.md index 525ff3b..be3dede 100644 --- a/docs/Chapter_19_Training_Data_Poisoning.md +++ b/docs/Chapter_19_Training_Data_Poisoning.md @@ -12,7 +12,7 @@ Related: Chapter 26 (Supply Chain), Chapter 30 (Backdoors) # Chapter 19: Training Data Poisoning

- +

_This chapter provides comprehensive coverage of training data poisoning attacks, backdoor injection techniques, model integrity compromise, detection methodologies, and defense strategies for LLM systems._ diff --git a/docs/Chapter_20_Model_Theft_and_Membership_Inference.md b/docs/Chapter_20_Model_Theft_and_Membership_Inference.md index 095d868..009ad3a 100644 --- a/docs/Chapter_20_Model_Theft_and_Membership_Inference.md +++ b/docs/Chapter_20_Model_Theft_and_Membership_Inference.md @@ -12,7 +12,7 @@ Related: Chapter 19 (Poisoning), Chapter 37 (Remediation) # Chapter 20: Model Theft and Membership Inference

- +

_This chapter provides comprehensive coverage of model extraction attacks, membership inference techniques, privacy violations in ML systems, intellectual property theft, watermarking, detection methods, and defense strategies for protecting model confidentiality._ diff --git a/docs/Chapter_21_Model_DoS_Resource_Exhaustion.md b/docs/Chapter_21_Model_DoS_Resource_Exhaustion.md index 5bf5e77..e377dda 100644 --- a/docs/Chapter_21_Model_DoS_Resource_Exhaustion.md +++ b/docs/Chapter_21_Model_DoS_Resource_Exhaustion.md @@ -12,7 +12,7 @@ Related: Chapter 26 (Supply Chain), Chapter 20 (Model Theft) # Chapter 21: Model DoS and Resource Exhaustion

- +

_This chapter covers Denial of Service (DoS) attacks on LLM systems, resource exhaustion techniques, economic attacks, detection methods, and defense strategies for protecting API availability and cost management._ diff --git a/docs/Chapter_22_Cross_Modal_Multimodal_Attacks.md b/docs/Chapter_22_Cross_Modal_Multimodal_Attacks.md index 0dc07cf..7eeb09a 100644 --- a/docs/Chapter_22_Cross_Modal_Multimodal_Attacks.md +++ b/docs/Chapter_22_Cross_Modal_Multimodal_Attacks.md @@ -12,7 +12,7 @@ Related: Chapter 25 (Adversarial ML), Chapter 21 (DoS) # Chapter 22: Cross-Modal and Multimodal Attacks

- +

_This chapter provides comprehensive coverage of attacks on multimodal AI systems, including vision-language models (GPT-4V, Claude 3, Gemini), image-based prompt injection, adversarial images, audio attacks, cross-modal exploitation techniques, detection methods, and defense strategies._ diff --git a/docs/Chapter_23_Advanced_Persistence_Chaining.md b/docs/Chapter_23_Advanced_Persistence_Chaining.md index 02ff515..0ed71b9 100644 --- a/docs/Chapter_23_Advanced_Persistence_Chaining.md +++ b/docs/Chapter_23_Advanced_Persistence_Chaining.md @@ -12,7 +12,7 @@ Related: Chapter 24 (Social Engineering), Chapter 26 (Autonomous Agents) # Chapter 23: Advanced Persistence and Chaining

- +

_This chapter provides comprehensive coverage of advanced persistence techniques and attack chaining for LLM systems, including context manipulation, multi-turn attacks, state persistence, chain-of-thought exploitation, prompt chaining, session hijacking, detection methods, and defense strategies._ diff --git a/docs/Chapter_24_Social_Engineering_LLMs.md b/docs/Chapter_24_Social_Engineering_LLMs.md index fc5ab78..dd355cd 100644 --- a/docs/Chapter_24_Social_Engineering_LLMs.md +++ b/docs/Chapter_24_Social_Engineering_LLMs.md @@ -12,7 +12,7 @@ Related: Chapter 23 (Persistence), Chapter 26 (Agents) # Chapter 24: Social Engineering with LLMs

- +

_This chapter provides comprehensive coverage of social engineering attacks powered by Large Language Models, including AI-generated phishing, impersonation attacks, trust exploitation, persuasion technique automation, spear phishing at scale, pretexting, detection methods, defense strategies, and critical ethical considerations._ diff --git a/docs/Chapter_25_Advanced_Adversarial_ML.md b/docs/Chapter_25_Advanced_Adversarial_ML.md index 224b4b8..2e02068 100644 --- a/docs/Chapter_25_Advanced_Adversarial_ML.md +++ b/docs/Chapter_25_Advanced_Adversarial_ML.md @@ -12,7 +12,7 @@ Related: Chapters 13 (Supply Chain Security), 20 (Model Extraction), 21 (Members # Chapter 25: Advanced Adversarial ML

- +

_This chapter digs into advanced adversarial machine learning, the kind of techniques that actually keep AI security researchers up at night. We'll cover gradient-based attacks, transferable adversarial examples, universal perturbations, model inversion, and (the big one) adversarial prompt optimization. You'll walk away understanding both how to use these techniques in authorized red team assessments and how to defend against them._ diff --git a/docs/Chapter_26_Supply_Chain_Attacks_on_AI.md b/docs/Chapter_26_Supply_Chain_Attacks_on_AI.md index 2096ea4..ad55b7a 100644 --- a/docs/Chapter_26_Supply_Chain_Attacks_on_AI.md +++ b/docs/Chapter_26_Supply_Chain_Attacks_on_AI.md @@ -12,7 +12,7 @@ Related: Chapters 13 (Supply Chain Security), 20 (Model Theft), 25 (Adversarial # Chapter 26: Supply Chain Attacks on AI

- +

_This chapter covers supply chain attacks targeting AI/ML systems: model repository compromises, dependency poisoning, malicious pre-trained models, compromised training pipelines, third-party API exploitation, plus detection methods, defense strategies, and ethical considerations for authorized testing._ diff --git a/docs/Chapter_27_Federated_Learning_Attacks.md b/docs/Chapter_27_Federated_Learning_Attacks.md index 9184094..371a1b4 100644 --- a/docs/Chapter_27_Federated_Learning_Attacks.md +++ b/docs/Chapter_27_Federated_Learning_Attacks.md @@ -12,7 +12,7 @@ Related: Chapters 11, 15, 19, 23 # Chapter 27: Federated Learning Attacks

- +

_Federated learning lets organizations train models together without sharing raw data. That's the promise, anyway. This chapter digs into why that promise is harder to keep than it sounds: model poisoning, gradient inversion, Byzantine failures, and the surprisingly difficult task of detecting when something's gone wrong. We'll cover attacks, defenses, and the ethical guardrails you need for legitimate security testing._ diff --git a/docs/Chapter_28_AI_Privacy_Attacks.md b/docs/Chapter_28_AI_Privacy_Attacks.md index 04a88d0..74dc62d 100644 --- a/docs/Chapter_28_AI_Privacy_Attacks.md +++ b/docs/Chapter_28_AI_Privacy_Attacks.md @@ -12,7 +12,7 @@ Related: Chapters 29 (Model Inversion), 27 (Federated Learning) # Chapter 28: AI Privacy Attacks

- +

_This chapter provides comprehensive coverage of AI privacy attacks, including Membership Inference Attacks (MIA), Attribute Inference, and Training Data Extraction. We explore how models unintentionally memorize and leak sensitive information, demonstrate practical extraction techniques, and outline robust defense strategies like Differential Privacy and Machine Unlearning to protect user data._ diff --git a/docs/Chapter_29_Model_Inversion_Attacks.md b/docs/Chapter_29_Model_Inversion_Attacks.md index d160a2d..fa8f16a 100644 --- a/docs/Chapter_29_Model_Inversion_Attacks.md +++ b/docs/Chapter_29_Model_Inversion_Attacks.md @@ -12,7 +12,7 @@ Related: Chapters 27 (Federated Learning), 25 (Adversarial ML) # Chapter 29: Model Inversion Attacks

- +

_This chapter explores Model Inversion Attacks (MIA), where adversaries reconstruct representative features or exact images of target classes by interrogating a model. We delve into optimization-based inversion, generative model priors, and the thin line between model utility and data leakage._ diff --git a/docs/Chapter_30_Backdoor_Attacks.md b/docs/Chapter_30_Backdoor_Attacks.md index d101f3b..fefda8d 100644 --- a/docs/Chapter_30_Backdoor_Attacks.md +++ b/docs/Chapter_30_Backdoor_Attacks.md @@ -12,7 +12,7 @@ Related: Chapters 27 (Federated Learning), 34 (Defense Evasion) # Chapter 30: Backdoor Attacks

- +

_This chapter dives deep into Backdoor Attacks (Trojans) in AI systems. Unlike standard adversarial examples that exploit existing fragility, backdoors are intentionally implanted vulnerabilities that trigger specific, malicious behaviors only when a secret "trigger" appears in the input. We cover injection mechanisms, clean-label attacks, Neural Cleanse detection, activation clustering defenses, supply chain vulnerabilities, and real-world case studies—including the $60 Wikipedia poisoning attack._ diff --git a/docs/Chapter_31_AI_System_Reconnaissance.md b/docs/Chapter_31_AI_System_Reconnaissance.md index aa56c14..90e172e 100644 --- a/docs/Chapter_31_AI_System_Reconnaissance.md +++ b/docs/Chapter_31_AI_System_Reconnaissance.md @@ -12,7 +12,7 @@ Related: Chapters 7 (Lab Setup), 32 (Automated Frameworks) # Chapter 31: AI System Reconnaissance

- +

_This chapter details the methodology for mapping the attack surface of AI/LLM deployments. We explore techniques for model fingerprinting, identifying backend infrastructure (Vector DBs, Orchestrators), and enumerating exposed APIs. It serves as the "Nmap" phase of AI Red Teaming._ diff --git a/docs/Chapter_32_Automated_Attack_Frameworks.md b/docs/Chapter_32_Automated_Attack_Frameworks.md index 21e6653..2273cb5 100644 --- a/docs/Chapter_32_Automated_Attack_Frameworks.md +++ b/docs/Chapter_32_Automated_Attack_Frameworks.md @@ -12,7 +12,7 @@ Related: Chapter 33 (Red Team Automation), Chapter 5 (Threat Modeling) # Chapter 32: Automated Attack Frameworks

- +

_This chapter provides comprehensive coverage of automated red teaming, detailing how to move from manual probing to industrial-scale vulnerability scanning. We explore the architecture of modular fuzzing harnesses, implement a custom generator-mutator-judge framework in Python, and analyze real-world incidents like the GCG attack to understand why automation is critical for uncovering deep adversarial flaws._ diff --git a/docs/Chapter_33_Red_Team_Automation.md b/docs/Chapter_33_Red_Team_Automation.md index d3404fd..418caed 100644 --- a/docs/Chapter_33_Red_Team_Automation.md +++ b/docs/Chapter_33_Red_Team_Automation.md @@ -12,7 +12,7 @@ Related: Chapters 38 (Continuous Red Teaming), 23 (LLM Evaluation) # Chapter 33: Red Team Automation

- +

_This chapter transitions from running ad-hoc tools to building continuous security pipelines (DevSecOps for AI). We demonstrate how to integrate fuzzers into GitHub Actions, define pass/fail thresholds for pull requests, and automate the reporting of security regression bugs in LLM applications._ diff --git a/docs/Chapter_34_Defense_Evasion_Techniques.md b/docs/Chapter_34_Defense_Evasion_Techniques.md index 6dd3d5c..2fe8d03 100644 --- a/docs/Chapter_34_Defense_Evasion_Techniques.md +++ b/docs/Chapter_34_Defense_Evasion_Techniques.md @@ -12,7 +12,7 @@ Related: Chapters 30 (Backdoors), 35 (Post-Exploitation) # Chapter 34: Defense Evasion Techniques

- +

_This chapter details the mechanisms attackers use to bypass AI security controls. We cover payload splitting, context flooding, obfuscation, and specialized encoding techniques designed to evade standard input filters and detection models._ diff --git a/docs/Chapter_35_Post-Exploitation_in_AI_Systems.md b/docs/Chapter_35_Post-Exploitation_in_AI_Systems.md index 7f6e794..7e07a1f 100644 --- a/docs/Chapter_35_Post-Exploitation_in_AI_Systems.md +++ b/docs/Chapter_35_Post-Exploitation_in_AI_Systems.md @@ -12,7 +12,7 @@ Related: Chapters 9 (LLM Components), 11 (Vulnerability Taxonomy) # Chapter 35: Post-Exploitation in AI Systems

- +

_This chapter focuses on the "what comes next" after a successful jailbreak or injection. We explore how compromised AI systems serve as landing pads for persistence, privilege escalation, and lateral movement into the broader cloud infrastructure, effectively turning the LLM into a Command and Control (C2) channel._ diff --git a/docs/Chapter_36_Reporting_and_Communication.md b/docs/Chapter_36_Reporting_and_Communication.md index 73d641e..fa37238 100644 --- a/docs/Chapter_36_Reporting_and_Communication.md +++ b/docs/Chapter_36_Reporting_and_Communication.md @@ -12,7 +12,7 @@ Related: Chapters 37 (Remediation), 38 (Continuous Red Teaming), 45 (Program Bui # Chapter 36: Reporting and Communication

- +

_This chapter teaches you how to craft impactful AI red team reports that communicate both technical and business risks effectively. You'll learn audience-tailored communication strategies, evidence-driven reporting structures, executive summary templates, detailed finding formats, and professional handoff procedures that ensure your findings drive actual security improvements._ diff --git a/docs/Chapter_37_Remediation_Strategies.md b/docs/Chapter_37_Remediation_Strategies.md index 54c3ed9..434bf2f 100644 --- a/docs/Chapter_37_Remediation_Strategies.md +++ b/docs/Chapter_37_Remediation_Strategies.md @@ -12,7 +12,7 @@ Related: Chapters 2 (Ethics/Communication), 4 (SOW/RoE), 38 (Continuous Red Team # Chapter 37: Presenting Results and Remediation Guidance

- +

_This chapter bridges technical findings and organizational action through effective presentation and remediation strategies. You'll learn how to present results to technical and non-technical audiences, facilitate collaborative remediation planning, create prioritized roadmaps, handle difficult conversations, and ensure your red teaming work translates into measurable security improvements._ diff --git a/docs/Chapter_38_Continuous_Red_Teaming.md b/docs/Chapter_38_Continuous_Red_Teaming.md index f23b620..38fe46e 100644 --- a/docs/Chapter_38_Continuous_Red_Teaming.md +++ b/docs/Chapter_38_Continuous_Red_Teaming.md @@ -12,7 +12,7 @@ Related: Chapters 45 (Program Building), 5 (Threat Modeling), 7 (Lab Setup) # Chapter 38: Continuous Red Teaming and Program Maturity

- +

_This chapter establishes a framework for continuous improvement and program maturity in AI red teaming. You'll learn common pitfalls to avoid, best practices for building effective red teaming capabilities, strategies for institutionalizing red teaming within organizations, and how to adapt your practice to the evolving AI threat landscape._ diff --git a/docs/Chapter_39_AI_Bug_Bounty_Programs.md b/docs/Chapter_39_AI_Bug_Bounty_Programs.md index 6847542..c687d5f 100644 --- a/docs/Chapter_39_AI_Bug_Bounty_Programs.md +++ b/docs/Chapter_39_AI_Bug_Bounty_Programs.md @@ -12,7 +12,7 @@ Related: Chapters 36 (Reporting), 40 (Compliance) # Chapter 39: AI Bug Bounty Programs

- +

_This chapter transforms the "dark art" of AI bug hunting into a rigorous engineering discipline. We move beyond manual prompt bashing to explore automated reconnaissance, traffic analysis, and the monetization of novel AI vulnerabilities._ diff --git a/docs/Chapter_40_Compliance_and_Standards.md b/docs/Chapter_40_Compliance_and_Standards.md index 32f88c4..0909436 100644 --- a/docs/Chapter_40_Compliance_and_Standards.md +++ b/docs/Chapter_40_Compliance_and_Standards.md @@ -12,7 +12,7 @@ Related: Chapter 02 (Ethics and Legal), Chapter 39 (Bug Bounty) # Chapter 40: Compliance and Standards

- +

In the enterprise, Red Teaming often means "Compliance Validation." This chapter turns abstract regulations—like the EU AI Act and ISO 42001—into concrete, testable engineering requirements. We will build tools to automatically audit AI systems against these legal frameworks. diff --git a/docs/Chapter_41_Industry_Best_Practices.md b/docs/Chapter_41_Industry_Best_Practices.md index 411ffb2..5ad928c 100644 --- a/docs/Chapter_41_Industry_Best_Practices.md +++ b/docs/Chapter_41_Industry_Best_Practices.md @@ -12,7 +12,7 @@ Related: Chapter 40 (Compliance) # Chapter 41: Industry Best Practices

- +

Security isn't a feature. It's an architecture. This chapter moves beyond "tips and tricks" to blueprint a production-grade AI defense stack, including advanced input sanitization, token-aware rate limiting, and automated circuit breakers. diff --git a/docs/Chapter_42_Case_Studies_and_War_Stories.md b/docs/Chapter_42_Case_Studies_and_War_Stories.md index 7db35df..3f0cec4 100644 --- a/docs/Chapter_42_Case_Studies_and_War_Stories.md +++ b/docs/Chapter_42_Case_Studies_and_War_Stories.md @@ -12,7 +12,7 @@ Related: Chapter 40 (Compliance) # Chapter 42: Case Studies and War Stories

- +

Analyzing failure is the foundation of security. We move beyond headlines to perform "Forensic Reconstruction" of major AI security incidents, revealing the specific code and architectural flaws that caused them. diff --git a/docs/Chapter_43_Future_of_AI_Red_Teaming.md b/docs/Chapter_43_Future_of_AI_Red_Teaming.md index 9702273..95a35f6 100644 --- a/docs/Chapter_43_Future_of_AI_Red_Teaming.md +++ b/docs/Chapter_43_Future_of_AI_Red_Teaming.md @@ -12,7 +12,7 @@ Related: Chapter 44 (Emerging Threats) # Chapter 43: Future of AI Red Teaming

- +

The era of "manual jailbreaking" is over. The future of AI Red Teaming is algorithmic, agentic, and autonomous. We explore the cutting-edge research—from Gradient-Based Optimization to Mechanistic Interpretability—that defines the next generation of security. diff --git a/docs/Chapter_44_Emerging_Threats.md b/docs/Chapter_44_Emerging_Threats.md index f1b9262..daf77df 100644 --- a/docs/Chapter_44_Emerging_Threats.md +++ b/docs/Chapter_44_Emerging_Threats.md @@ -12,7 +12,7 @@ Related: Chapter 43 (Future of AI) # Chapter 44: Emerging Threats

- +

The threat landscape is bigger than simple chatbot hacks. This chapter provides the technical tools to detect "Shadow AI" on your network, analyze the risks of "Audio Adversarial Attacks," and prevent "Log Injection" in Critical Infrastructure. diff --git a/docs/Chapter_45_Building_an_AI_Red_Team_Program.md b/docs/Chapter_45_Building_an_AI_Red_Team_Program.md index eec8b7b..85d93c2 100644 --- a/docs/Chapter_45_Building_an_AI_Red_Team_Program.md +++ b/docs/Chapter_45_Building_an_AI_Red_Team_Program.md @@ -12,7 +12,7 @@ Related: Chapter 38 (Continuous Red Teaming) # Chapter 45: Building an AI Red Team Program

- +

Turning the Red Team mindset into operations takes more than hacking skills. It requires a formal program, a budget, and a defined scope. This chapter provides the blueprint for CISOs and Directors to build an in-house AI Security capability. diff --git a/docs/Chapter_46_Conclusion_and_Next_Steps.md b/docs/Chapter_46_Conclusion_and_Next_Steps.md index 05e8496..3ca56ac 100644 --- a/docs/Chapter_46_Conclusion_and_Next_Steps.md +++ b/docs/Chapter_46_Conclusion_and_Next_Steps.md @@ -11,7 +11,7 @@ Prerequisites: None # Chapter 46: Conclusion and Next Steps

- +

_You've reached the end of the AI LLM Red Team Handbook. But in security, there is no end—only the next model update._ diff --git a/docs/archive/AI LLM Red Team Handbook.md b/docs/archive/AI LLM Red Team Handbook.md index b0fd9c4..9e6a7bc 100644 --- a/docs/archive/AI LLM Red Team Handbook.md +++ b/docs/archive/AI LLM Red Team Handbook.md @@ -89,7 +89,7 @@ Related: Chapters 2 (Ethics), 3 (Mindset), 7 (Lab Setup) # Chapter 1: Introduction to AI Red Teaming -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter introduces AI red teaming as a structured practice for identifying vulnerabilities in AI systems, particularly Large Language Models. You'll learn what makes AI security different from traditional cybersecurity, understand the unique attack surfaces of LLMs, and explore the professional mindset and ethical framework required for responsible AI security testing._ @@ -319,7 +319,7 @@ Related: Chapters 4 (SOW/RoE), 8 (Evidence), 36 (Reporting) # Chapter 2: Ethics, Legal, and Stakeholder Communication -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes the critical ethical and legal framework for AI red teaming. You'll learn the principles of responsible security testing, understand legal boundaries and authorization requirements, master stakeholder communication strategies, and develop professional practices for data handling and responsible disclosure in AI security engagements._ @@ -551,7 +551,7 @@ Related: Chapters 2 (Ethics), 5 (Threat Modeling) # Chapter 3: The Red Teamer's Mindset -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter develops the critical thinking skills and adversarial perspective essential for effective AI red teaming. You'll learn how to think like an attacker, cultivate skepticism and creativity, balance persistence with professionalism, and develop the psychological resilience needed for this specialized security discipline._ @@ -775,7 +775,7 @@ Related: Chapters 2 (Ethics), 6 (Scoping), 8 (Evidence) # Chapter 4: SOW, Rules of Engagement, and Client Onboarding -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides practical frameworks for establishing professional AI red team engagements. You'll learn to create comprehensive Statements of Work, define clear Rules of Engagement, conduct effective client onboarding, and establish communication protocols that ensure legal protection and professional success._ @@ -1050,7 +1050,7 @@ Related: Chapters 6 (Scoping), 9 (Architectures), 31 (Reconnaissance) # Chapter 5: Threat Modeling and Risk Analysis -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter teaches systematic threat modeling and risk analysis for AI systems. You'll learn to identify critical assets, enumerate threat actors, map attack surfaces specific to LLMs, prioritize risks by impact and likelihood, and create actionable threat models that drive effective red team engagements._ @@ -1229,7 +1229,7 @@ Related: Chapters 4 (SOW/RoE), 5 (Threat Modeling), 7 (Lab Setup) # Chapter 6: Scoping an Engagement -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter guides you through the critical process of scoping AI red team engagements. You'll learn to define realistic boundaries, estimate timelines and resources, identify technical and business constraints, communicate scope effectively to stakeholders, and avoid common scoping pitfalls that derail engagements._ @@ -1382,7 +1382,7 @@ Related: Chapters 9 (Architectures), 32 (Automation), 33 (Red Team Frameworks) # Chapter 7: Lab Setup and Environmental Safety -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides hands-on guidance for setting up safe, isolated AI red teaming environments. You'll learn to configure local and cloud-based labs, implement proper network isolation, deploy test models and applications, establish monitoring and logging, and create reproducible test environments for ethical AI security research._ @@ -1524,7 +1524,7 @@ Related: Chapters 2 (Ethics), 36 (Reporting), 40 (Compliance) # Chapter 8: Evidence, Documentation, and Chain of Custody -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes professional standards for evidence collection and documentation in AI red team engagements. You'll learn forensically-sound capture methods, proper chain-of-custody procedures, secure evidence storage and handling, comprehensive documentation practices, and legal requirements for preserving findings that may be used in compliance or legal contexts._ @@ -1685,7 +1685,7 @@ Related: Chapters 10 (Tokenization), 11 (Plugins), 12 (RAG) # Chapter 9: LLM Architectures and System Components -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides deep technical understanding of LLM architectures critical for red teaming. You'll learn transformer internals, attention mechanisms, embedding spaces, system prompts and context windows, model deployment patterns, and how architectural choices create specific attack surfaces that red teamers can exploit._ @@ -1872,7 +1872,7 @@ Related: Chapters 9 (Architectures), 14 (Prompt Injection), 18 (Evasion) # Chapter 10: Tokenization, Context, and Generation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter explores how LLMs process and generate text, with focus on security implications. You'll learn tokenization mechanisms (BPE, WordPiece), context window management, generation strategies (greedy, sampling, beam search), and how understanding these processes enables sophisticated attacks like token manipulation and evasion techniques._ @@ -2024,7 +2024,7 @@ Related: Chapters 12 (RAG), 17 (Plugin Exploitation), 26 (Supply Chain) # Chapter 11: Plugins, Extensions, and External APIs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter examines the plugin and API ecosystem that extends LLM capabilities and creates new attack surfaces. You'll learn plugin architectures, function calling mechanisms, API integration patterns, authentication and authorization flows, and the unique vulnerabilities introduced when LLMs orchestrate external tool usage._ @@ -2185,7 +2185,7 @@ Related: Chapters 11 (Plugins), 13 (Supply Chain), 23 (Persistence) # Chapter 12: Retrieval-Augmented Generation (RAG) Pipelines -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter dissects Retrieval Augmented Generation systems and their attack surfaces. You'll learn RAG architecture (indexing, embedding, retrieval, generation), vector database security, context injection through retrieval poisoning, prompt leakage via retrieved documents, and how to test the complex data flow that makes RAG both powerful and vulnerable._ @@ -3494,7 +3494,7 @@ Related: Chapters 11 (Plugins), 19 (Training Data Poisoning), 26 (Supply Chain A # Chapter 13: Data Provenance and Supply Chain Security -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter addresses the critical but often overlooked aspect of AI supply chain security. You'll learn to trace data and model provenance, identify supply chain attack surfaces (datasets, pre-trained models, dependencies), assess third-party components, verify model integrity, and establish security controls that protect against poisoned training data and compromised model artifacts._ @@ -5384,7 +5384,7 @@ Related: Chapters 15-16 (Data Leakage, Jailbreaks) # Chapter 14: Prompt Injection (Direct/Indirect, 1st/3rd Party) -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of prompt injection attacks, including direct and indirect injection techniques, first-party and third-party variations, detection methods, defense-in-depth strategies, real-world case studies, and critical ethical considerations for authorized security testing._ @@ -9647,7 +9647,7 @@ Related: Chapters 16, 20 (Jailbreaks, Model Theft) # Chapter 15: Data Leakage and Extraction -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of data leakage vulnerabilities in LLM systems, including training data extraction, conversation history leakage, system prompt disclosure, credential extraction, PII revelation, model inversion attacks, detection strategies, mitigation techniques, and critical regulatory compliance considerations._ @@ -13533,7 +13533,7 @@ Related: Chapters 11, 17 (Safety/Alignment, Plugin Exploitation) # Chapter 16: Jailbreaks and Bypass Techniques -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of jailbreak techniques, bypass methods, testing methodologies, and defenses for LLM systems._ @@ -15132,7 +15132,7 @@ Related: Chapter 15 (Data Leakage), Chapter 23 (Persistence) # Chapter 17: Plugin and API Exploitation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of security issues in LLM plugins, APIs, and third-party integrations, including architecture analysis, vulnerability discovery, exploitation techniques, and defensive strategies._ @@ -17943,7 +17943,7 @@ Related: Chapter 16 (Jailbreaks), Chapter 21 (DoS) # Chapter 18: Evasion, Obfuscation, and Adversarial Inputs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of evasion techniques, obfuscation methods, and adversarial input strategies used to bypass LLM security controls, along with detection and mitigation approaches._ @@ -19208,7 +19208,7 @@ Related: Chapter 26 (Supply Chain), Chapter 30 (Backdoors) # Chapter 19: Training Data Poisoning -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of training data poisoning attacks, backdoor injection techniques, model integrity compromise, detection methodologies, and defense strategies for LLM systems._ @@ -20172,7 +20172,7 @@ Related: Chapter 19 (Poisoning), Chapter 37 (Remediation) # Chapter 20: Model Theft and Membership Inference -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of model extraction attacks, membership inference techniques, privacy violations in ML systems, intellectual property theft, watermarking, detection methods, and defense strategies for protecting model confidentiality._ @@ -21231,7 +21231,7 @@ Related: Chapter 26 (Supply Chain), Chapter 20 (Model Theft) # Chapter 21: Model DoS and Resource Exhaustion -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of Denial of Service (DoS) attacks on LLM systems, resource exhaustion techniques, economic attacks, detection methods, and defense strategies for protecting API availability and cost management._ @@ -22135,7 +22135,7 @@ Related: Chapter 25 (Adversarial ML), Chapter 21 (DoS) # Chapter 22: Cross-Modal and Multimodal Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of attacks on multimodal AI systems, including vision-language models (GPT-4V, Claude 3, Gemini), image-based prompt injection, adversarial images, audio attacks, cross-modal exploitation techniques, detection methods, and defense strategies._ @@ -23436,7 +23436,7 @@ Related: Chapter 24 (Social Engineering), Chapter 26 (Autonomous Agents) # Chapter 23: Advanced Persistence and Chaining -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of advanced persistence techniques and attack chaining for LLM systems, including context manipulation, multi-turn attacks, state persistence, chain-of-thought exploitation, prompt chaining, session hijacking, detection methods, and defense strategies._ @@ -24333,7 +24333,7 @@ Related: Chapter 23 (Persistence), Chapter 26 (Agents) # Chapter 24: Social Engineering with LLMs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of social engineering attacks powered by Large Language Models, including AI-generated phishing, impersonation attacks, trust exploitation, persuasion technique automation, spear phishing at scale, pretexting, detection methods, defense strategies, and critical ethical considerations._ @@ -25477,7 +25477,7 @@ Related: Chapters 13 (Supply Chain Security), 20 (Model Extraction), 21 (Members # Chapter 25: Advanced Adversarial ML -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter digs into advanced adversarial machine learning, the kind of techniques that actually keep AI security researchers up at night. We'll cover gradient-based attacks, transferable adversarial examples, universal perturbations, model inversion, and (the big one) adversarial prompt optimization. You'll walk away understanding both how to use these techniques in authorized red team assessments and how to defend against them._ @@ -26701,7 +26701,7 @@ Related: Chapters 13 (Supply Chain Security), 20 (Model Theft), 25 (Adversarial # Chapter 26: Supply Chain Attacks on AI -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter covers supply chain attacks targeting AI/ML systems: model repository compromises, dependency poisoning, malicious pre-trained models, compromised training pipelines, third-party API exploitation, plus detection methods, defense strategies, and ethical considerations for authorized testing._ @@ -27523,7 +27523,7 @@ Supply chain attacks compromise AI by injecting malicious code, backdoors, or po --- # Chapter 27: Federated Learning Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27540,7 +27540,7 @@ Content for this chapter will be added in future updates. --- # Chapter 28: AI Privacy Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27557,7 +27557,7 @@ Content for this chapter will be added in future updates. --- # Chapter 29: Model Inversion Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27574,7 +27574,7 @@ Content for this chapter will be added in future updates. --- # Chapter 30: Backdoor Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27591,7 +27591,7 @@ Content for this chapter will be added in future updates. --- # Chapter 31: AI System Reconnaissance -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27608,7 +27608,7 @@ Content for this chapter will be added in future updates. --- # Chapter 32: Automated Attack Frameworks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27625,7 +27625,7 @@ Content for this chapter will be added in future updates. --- # Chapter 33: Red Team Automation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27642,7 +27642,7 @@ Content for this chapter will be added in future updates. --- # Chapter 34: Defense Evasion Techniques -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27659,7 +27659,7 @@ Content for this chapter will be added in future updates. --- # Chapter 35: Post-Exploitation in AI Systems -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27687,7 +27687,7 @@ Related: Chapters 37 (Remediation), 38 (Continuous Red Teaming), 45 (Program Bui # Chapter 36: Reporting and Communication -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter teaches you how to craft impactful AI red team reports that communicate both technical and business risks effectively. You'll learn audience-tailored communication strategies, evidence-driven reporting structures, executive summary templates, detailed finding formats, and professional handoff procedures that ensure your findings drive actual security improvements._ @@ -27863,7 +27863,7 @@ Related: Chapters 2 (Ethics/Communication), 4 (SOW/RoE), 38 (Continuous Red Team # Chapter 37: Presenting Results and Remediation Guidance -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter bridges technical findings and organizational action through effective presentation and remediation strategies. You'll learn how to present results to technical and non-technical audiences, facilitate collaborative remediation planning, create prioritized roadmaps, handle difficult conversations, and ensure your red teaming work translates into measurable security improvements._ @@ -27999,7 +27999,7 @@ Related: Chapters 45 (Program Building), 5 (Threat Modeling), 7 (Lab Setup) # Chapter 38: Lessons Learned and Program Maturity -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes a framework for continuous improvement and program maturity in AI red teaming. You'll learn common pitfalls to avoid, best practices for building effective red teaming capabilities, strategies for institutionalizing red teaming within organizations, and how to adapt your practice to the evolving AI threat landscape._ @@ -28099,7 +28099,7 @@ To make AI red teaming a sustainable part of your organization’s security post With lessons learned and program maturity frameworks in place, Chapter 45 provides a comprehensive blueprint for building world-class AI red team programs - covering team structure, skill sets, engagement lifecycles, and the evolution from tactical assessments to strategic wargaming. # Chapter 39: AI Bug Bounty Programs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28116,7 +28116,7 @@ Content for this chapter will be added in future updates. --- # Chapter 40: Compliance and Standards -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28133,7 +28133,7 @@ Content for this chapter will be added in future updates. --- # Chapter 41: Industry Best Practices -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28150,7 +28150,7 @@ Content for this chapter will be added in future updates. --- # Chapter 42: Case Studies and War Stories -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28167,7 +28167,7 @@ Content for this chapter will be added in future updates. --- # Chapter 43: Future of AI Red Teaming -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28184,7 +28184,7 @@ Content for this chapter will be added in future updates. --- # Chapter 44: Emerging Threats -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -28212,7 +28212,7 @@ Related: Chapters 1-4 (Foundations), 38 (Program Maturity), 36-37 (Ops) # Chapter 45: Building an AI Red Team Program -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides a comprehensive blueprint for establishing world-class AI red team programs. You'll learn organizational models, essential skill sets, the adversarial mindset, engagement lifecycles, tool arsenals, vulnerability taxonomies, integration strategies, and the evolution from tactical assessments to strategic wargaming._ @@ -28506,7 +28506,7 @@ A mature AI red team anticipates future threats and drives systemic improvement - Measure and demonstrate value through tangible risk reduction and improved resilience. # Chapter 46: Conclusion and Next Steps -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ diff --git a/docs/archive/AI LLM Red Team Handbook_new.md b/docs/archive/AI LLM Red Team Handbook_new.md index ac6b086..2f8d923 100644 --- a/docs/archive/AI LLM Red Team Handbook_new.md +++ b/docs/archive/AI LLM Red Team Handbook_new.md @@ -78,7 +78,7 @@ ## Chapter 1: Introduction to AI Red Teaming -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter introduces AI red teaming as a structured practice for identifying vulnerabilities in AI systems, particularly Large Language Models. You'll learn what makes AI security different from traditional cybersecurity, understand the unique attack surfaces of LLMs, and explore the professional mindset and ethical framework required for responsible AI security testing._ @@ -298,7 +298,7 @@ _Proceed to Chapter 2 to explore ethical and legal essentials in depth, and begi ## Chapter 2: Ethics, Legal, and Stakeholder Communication -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes the critical ethical and legal framework for AI red teaming. You'll learn the principles of responsible security testing, understand legal boundaries and authorization requirements, master stakeholder communication strategies, and develop professional practices for data handling and responsible disclosure in AI security engagements._ @@ -520,7 +520,7 @@ _In the next chapter, you'll develop the mindset that distinguishes effective AI ## Chapter 3: The Red Teamer's Mindset -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter develops the critical thinking skills and adversarial perspective essential for effective AI red teaming. You'll learn how to think like an attacker, cultivate skepticism and creativity, balance persistence with professionalism, and develop the psychological resilience needed for this specialized security discipline._ @@ -735,7 +735,7 @@ _Mastering the red team mindset primes you for the work ahead: scoping, planning ## Chapter 4: SOW, Rules of Engagement, and Client Onboarding -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides practical frameworks for establishing professional AI red team engagements. You'll learn to create comprehensive Statements of Work, define clear Rules of Engagement, conduct effective client onboarding, and establish communication protocols that ensure legal protection and professional success._ @@ -1000,7 +1000,7 @@ _Solid foundations prevent project failure and foster trust. The next chapter wi ## Chapter 5: Threat Modeling and Risk Analysis -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter teaches systematic threat modeling and risk analysis for AI systems. You'll learn to identify critical assets, enumerate threat actors, map attack surfaces specific to LLMs, prioritize risks by impact and likelihood, and create actionable threat models that drive effective red team engagements._ @@ -1169,7 +1169,7 @@ As AI systems become more complex with multi-modal capabilities, agentic behavio ## Chapter 6: Scoping an Engagement -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter guides you through the critical process of scoping AI red team engagements. You'll learn to define realistic boundaries, estimate timelines and resources, identify technical and business constraints, communicate scope effectively to stakeholders, and avoid common scoping pitfalls that derail engagements._ @@ -1312,7 +1312,7 @@ As AI systems grow in complexity and integration depth, scoping will become more ## Chapter 7: Lab Setup and Environmental Safety -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides hands-on guidance for setting up safe, isolated AI red teaming environments. You'll learn to configure local and cloud-based labs, implement proper network isolation, deploy test models and applications, establish monitoring and logging, and create reproducible test environments for ethical AI security research._ @@ -1444,7 +1444,7 @@ Expect to see standardized AI red teaming lab templates, infrastructure-as-code ## Chapter 8: Evidence, Documentation, and Chain of Custody -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes professional standards for evidence collection and documentation in AI red team engagements. You'll learn forensically-sound capture methods, proper chain-of-custody procedures, secure evidence storage and handling, comprehensive documentation practices, and legal requirements for preserving findings that may be used in compliance or legal contexts._ @@ -1595,7 +1595,7 @@ Expect to see automated evidence collection tools integrated into red team platf ## Chapter 9: LLM Architectures and System Components -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides deep technical understanding of LLM architectures critical for red teaming. You'll learn transformer internals, attention mechanisms, embedding spaces, system prompts and context windows, model deployment patterns, and how architectural choices create specific attack surfaces that red teamers can exploit._ @@ -1772,7 +1772,7 @@ As LLM architectures evolve with mixture-of-experts, sparse attention, and multi ## Chapter 10: Tokenization, Context, and Generation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter explores how LLMs process and generate text, with focus on security implications. You'll learn tokenization mechanisms (BPE, WordPiece), context window management, generation strategies (greedy, sampling, beam search), and how understanding these processes enables sophisticated attacks like token manipulation and evasion techniques._ @@ -1914,7 +1914,7 @@ Evolving tokenization approaches (character-level, byte-level, learned vocabular ## Chapter 11: Plugins, Extensions, and External APIs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter examines the plugin and API ecosystem that extends LLM capabilities and creates new attack surfaces. You'll learn plugin architectures, function calling mechanisms, API integration patterns, authentication and authorization flows, and the unique vulnerabilities introduced when LLMs orchestrate external tool usage._ @@ -2065,7 +2065,7 @@ As LLMs gain more agentic capabilities with tool use and multi-step planning, pl ## Chapter 12: Retrieval-Augmented Generation (RAG) Pipelines -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter dissects Retrieval Augmented Generation systems and their attack surfaces. You'll learn RAG architecture (indexing, embedding, retrieval, generation), vector database security, context injection through retrieval poisoning, prompt leakage via retrieved documents, and how to test the complex data flow that makes RAG both powerful and vulnerable._ @@ -3364,7 +3364,7 @@ As RAG systems become more sophisticated with multi-hop retrieval, cross-modal s ## Chapter 13: Data Provenance and Supply Chain Security -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter addresses the critical but often overlooked aspect of AI supply chain security. You'll learn to trace data and model provenance, identify supply chain attack surfaces (datasets, pre-trained models, dependencies), assess third-party components, verify model integrity, and establish security controls that protect against poisoned training data and compromised model artifacts._ @@ -5244,7 +5244,7 @@ As AI supply chains grow more complex with model marketplaces, federated learnin ## Chapter 14: Prompt Injection (Direct/Indirect, 1st/3rd Party) -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of prompt injection attacks, including direct and indirect injection techniques, first-party and third-party variations, detection methods, defense-in-depth strategies, real-world case studies, and critical ethical considerations for authorized security testing._ @@ -9497,7 +9497,7 @@ _Prompt injection represents the defining security challenge of the LLM era. Lik ## Chapter 15: Data Leakage and Extraction -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of data leakage vulnerabilities in LLM systems, including training data extraction, conversation history leakage, system prompt disclosure, credential extraction, PII revelation, model inversion attacks, detection strategies, mitigation techniques, and critical regulatory compliance considerations._ @@ -13373,7 +13373,7 @@ _Continue to Chapter 16: Jailbreaks and Bypass Techniques to learn how attackers ## Chapter 16: Jailbreaks and Bypass Techniques -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of jailbreak techniques, bypass methods, testing methodologies, and defenses for LLM systems._ @@ -14962,7 +14962,7 @@ Jailbreaks bypass LLM safety controls through role-playing, instruction manipula ## Chapter 17: Plugin and API Exploitation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of security issues in LLM plugins, APIs, and third-party integrations, including architecture analysis, vulnerability discovery, exploitation techniques, and defensive strategies._ @@ -17764,7 +17764,7 @@ Attackers manipulate the LLM to invoke plugins/APIs with malicious arguments or ## Chapter 18: Evasion, Obfuscation, and Adversarial Inputs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of evasion techniques, obfuscation methods, and adversarial input strategies used to bypass LLM security controls, along with detection and mitigation approaches._ @@ -19020,7 +19020,7 @@ Evasion attacks manipulate input prompts to bypass content filters and safety gu ## Chapter 19: Training Data Poisoning -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of training data poisoning attacks, backdoor injection techniques, model integrity compromise, detection methodologies, and defense strategies for LLM systems._ @@ -19975,7 +19975,7 @@ Attackers inject malicious data into the training set (pre-training or fine-tuni ## Chapter 20: Model Theft and Membership Inference -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of model extraction attacks, membership inference techniques, privacy violations in ML systems, intellectual property theft, watermarking, detection methods, and defense strategies for protecting model confidentiality._ @@ -21025,7 +21025,7 @@ Attackers query the model to either learn its internal parameters (Model Theft) ## Chapter 21: Model DoS and Resource Exhaustion -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of Denial of Service (DoS) attacks on LLM systems, resource exhaustion techniques, economic attacks, detection methods, and defense strategies for protecting API availability and cost management._ @@ -21920,7 +21920,7 @@ Attackers exploit the high computational and financial cost of LLM inference ($O ## Chapter 22: Cross-Modal and Multimodal Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of attacks on multimodal AI systems, including vision-language models (GPT-4V, Claude 3, Gemini), image-based prompt injection, adversarial images, audio attacks, cross-modal exploitation techniques, detection methods, and defense strategies._ @@ -23212,7 +23212,7 @@ Using non-text inputs (Images, Audio) to inject prompts or adversarial noise tha ## Chapter 23: Advanced Persistence and Chaining -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of advanced persistence techniques and attack chaining for LLM systems, including context manipulation, multi-turn attacks, state persistence, chain-of-thought exploitation, prompt chaining, session hijacking, detection methods, and defense strategies._ @@ -24100,7 +24100,7 @@ Attackers manipulate the model's "memory" (context window, RAG database, or syst ## Chapter 24: Social Engineering with LLMs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides comprehensive coverage of social engineering attacks powered by Large Language Models, including AI-generated phishing, impersonation attacks, trust exploitation, persuasion technique automation, spear phishing at scale, pretexting, detection methods, defense strategies, and critical ethical considerations._ @@ -25235,7 +25235,7 @@ Attackers leverage LLMs to automate the creation of highly personalized, persuas ## Chapter 25: Advanced Adversarial ML -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter digs into advanced adversarial machine learning, the kind of techniques that actually keep AI security researchers up at night. We'll cover gradient-based attacks, transferable adversarial examples, universal perturbations, model inversion, and (the big one) adversarial prompt optimization. You'll walk away understanding both how to use these techniques in authorized red team assessments and how to defend against them._ @@ -26449,7 +26449,7 @@ Advanced Adversarial ML attacks use mathematical optimization to find minimal pe ## Chapter 26: Supply Chain Attacks on AI -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter covers supply chain attacks targeting AI/ML systems: model repository compromises, dependency poisoning, malicious pre-trained models, compromised training pipelines, third-party API exploitation, plus detection methods, defense strategies, and ethical considerations for authorized testing._ @@ -27272,7 +27272,7 @@ Supply chain attacks compromise AI by injecting malicious code, backdoors, or po ## Chapter 27: Federated Learning Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27292,7 +27292,7 @@ Content for this chapter will be added in future updates. ## Chapter 28: AI Privacy Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27312,7 +27312,7 @@ Content for this chapter will be added in future updates. ## Chapter 29: Model Inversion Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27332,7 +27332,7 @@ Content for this chapter will be added in future updates. ## Chapter 30: Backdoor Attacks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27352,7 +27352,7 @@ Content for this chapter will be added in future updates. ## Chapter 31: AI System Reconnaissance -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27372,7 +27372,7 @@ Content for this chapter will be added in future updates. ## Chapter 32: Automated Attack Frameworks -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27392,7 +27392,7 @@ Content for this chapter will be added in future updates. ## Chapter 33: Red Team Automation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27412,7 +27412,7 @@ Content for this chapter will be added in future updates. ## Chapter 34: Defense Evasion Techniques -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27432,7 +27432,7 @@ Content for this chapter will be added in future updates. ## Chapter 35: Post-Exploitation in AI Systems -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27452,7 +27452,7 @@ Content for this chapter will be added in future updates. ## Chapter 36: Reporting and Communication -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter teaches you how to craft impactful AI red team reports that communicate both technical and business risks effectively. You'll learn audience-tailored communication strategies, evidence-driven reporting structures, executive summary templates, detailed finding formats, and professional handoff procedures that ensure your findings drive actual security improvements._ @@ -27618,7 +27618,7 @@ Chapter 37 covers presenting your results effectively to both technical and non- ## Chapter 37: Presenting Results and Remediation Guidance -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter bridges technical findings and organizational action through effective presentation and remediation strategies. You'll learn how to present results to technical and non-technical audiences, facilitate collaborative remediation planning, create prioritized roadmaps, handle difficult conversations, and ensure your red teaming work translates into measurable security improvements._ @@ -27744,7 +27744,7 @@ Chapter 38 explores lessons learned, common pitfalls, and strategies for buildin ## Chapter 38: Lessons Learned and Program Maturity -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter establishes a framework for continuous improvement and program maturity in AI red teaming. You'll learn common pitfalls to avoid, best practices for building effective red teaming capabilities, strategies for institutionalizing red teaming within organizations, and how to adapt your practice to the evolving AI threat landscape._ @@ -27845,7 +27845,7 @@ With lessons learned and program maturity frameworks in place, Chapter 45 provid ## Chapter 39: AI Bug Bounty Programs -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27865,7 +27865,7 @@ Content for this chapter will be added in future updates. ## Chapter 40: Compliance and Standards -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27885,7 +27885,7 @@ Content for this chapter will be added in future updates. ## Chapter 41: Industry Best Practices -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27905,7 +27905,7 @@ Content for this chapter will be added in future updates. ## Chapter 42: Case Studies and War Stories -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27925,7 +27925,7 @@ Content for this chapter will be added in future updates. ## Chapter 43: Future of AI Red Teaming -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27945,7 +27945,7 @@ Content for this chapter will be added in future updates. ## Chapter 44: Emerging Threats -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ @@ -27965,7 +27965,7 @@ Content for this chapter will be added in future updates. ## Chapter 45: Building an AI Red Team Program -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter provides a comprehensive blueprint for establishing world-class AI red team programs. You'll learn organizational models, essential skill sets, the adversarial mindset, engagement lifecycles, tool arsenals, vulnerability taxonomies, integration strategies, and the evolution from tactical assessments to strategic wargaming._ @@ -28260,7 +28260,7 @@ A mature AI red team anticipates future threats and drives systemic improvement ## Chapter 46: Conclusion and Next Steps -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter is currently under development._ diff --git a/docs/archive/AI_LLM_Red_Team_Field_Manual_backup_20251207.md b/docs/archive/AI_LLM_Red_Team_Field_Manual_backup_20251207.md index 1966e6b..02fabac 100644 --- a/docs/archive/AI_LLM_Red_Team_Field_Manual_backup_20251207.md +++ b/docs/archive/AI_LLM_Red_Team_Field_Manual_backup_20251207.md @@ -1,6 +1,6 @@ # **AI/LLM Red Team Field Manual** -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) > **For Junior Penetration Testers**: This manual is designed as a complete, standalone field guide. Follow the Quick Start below to begin testing within 15 minutes. @@ -186,7 +186,7 @@ For step-by-step attack procedures with extensive code examples, see the **modul --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **1. Introduction: Rules of Engagement (RoE)** @@ -204,7 +204,7 @@ Define in writing: in-scope systems/models, allowed techniques, test windows, ha --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **1.5 Environment Setup & Configuration** @@ -490,7 +490,7 @@ API Config File: ✅ --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **2. Red Teaming Phases** @@ -765,7 +765,7 @@ chmod +x cleanup.sh --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **3\. Attack Types & Practical Test Examples** @@ -1250,7 +1250,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@ --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **4\. Tools Reference & CLI Commands** @@ -1296,7 +1296,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@ --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **5\. Attack-Type–to–Tool Quick Lookup Table** @@ -1315,7 +1315,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@ --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **5.5 API Configuration Guide** @@ -1583,7 +1583,7 @@ echo "✅ Provider test complete!" --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **6\. Reporting Guidance** @@ -1597,7 +1597,7 @@ Report every finding with: --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **7\. Additional Guidance** @@ -1609,7 +1609,7 @@ Report every finding with: --- -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) ## **8. Troubleshooting Guide** diff --git a/docs/archive/Chapter_17_Plugin_and_API_Exploitation.md b/docs/archive/Chapter_17_Plugin_and_API_Exploitation.md index 412ac1c..694033c 100644 --- a/docs/archive/Chapter_17_Plugin_and_API_Exploitation.md +++ b/docs/archive/Chapter_17_Plugin_and_API_Exploitation.md @@ -11,7 +11,7 @@ Related: Chapter 15 (Data Leakage), Chapter 23 (Persistence) # Chapter 17: Plugin and API Exploitation -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _This chapter covers security issues in LLM plugins, APIs, and third-party integrations—from architecture analysis and vulnerability discovery to exploitation techniques and defense strategies._ diff --git a/docs/field_manuals/Field_Manual_01_Prompt_Injection_Playbook.md b/docs/field_manuals/Field_Manual_01_Prompt_Injection_Playbook.md index 5db3867..7c28126 100644 --- a/docs/field_manuals/Field_Manual_01_Prompt_Injection_Playbook.md +++ b/docs/field_manuals/Field_Manual_01_Prompt_Injection_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 01: Prompt Injection -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Prompt Injection (Direct & Indirect) **Difficulty**: ⭐ Beginner diff --git a/docs/field_manuals/Field_Manual_02_Data_Leakage_Playbook.md b/docs/field_manuals/Field_Manual_02_Data_Leakage_Playbook.md index 52edd61..693c573 100644 --- a/docs/field_manuals/Field_Manual_02_Data_Leakage_Playbook.md +++ b/docs/field_manuals/Field_Manual_02_Data_Leakage_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 02: Data Leakage & Extraction -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Training Data & Memory Extraction **Difficulty**: ⭐⭐ Intermediate diff --git a/docs/field_manuals/Field_Manual_03_Jailbreak_Playbook.md b/docs/field_manuals/Field_Manual_03_Jailbreak_Playbook.md index c5d9f6a..ef2d0bb 100644 --- a/docs/field_manuals/Field_Manual_03_Jailbreak_Playbook.md +++ b/docs/field_manuals/Field_Manual_03_Jailbreak_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 03: Jailbreaks & Bypass -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Content Filter & Safety Bypass **Difficulty**: ⭐ Beginner diff --git a/docs/field_manuals/Field_Manual_04_Plugin_Exploitation_Playbook.md b/docs/field_manuals/Field_Manual_04_Plugin_Exploitation_Playbook.md index 1009213..691d531 100644 --- a/docs/field_manuals/Field_Manual_04_Plugin_Exploitation_Playbook.md +++ b/docs/field_manuals/Field_Manual_04_Plugin_Exploitation_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 04: Plugin & API Exploitation -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Plugin/Function Calling Exploitation **Difficulty**: ⭐⭐⭐ Advanced diff --git a/docs/field_manuals/Field_Manual_05_Evasion_Playbook.md b/docs/field_manuals/Field_Manual_05_Evasion_Playbook.md index 7541508..8a99c1c 100644 --- a/docs/field_manuals/Field_Manual_05_Evasion_Playbook.md +++ b/docs/field_manuals/Field_Manual_05_Evasion_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 05: Evasion & Obfuscation -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Input Filter Bypass **Difficulty**: ⭐⭐ Intermediate diff --git a/docs/field_manuals/Field_Manual_06_Data_Poisoning_Playbook.md b/docs/field_manuals/Field_Manual_06_Data_Poisoning_Playbook.md index b329eea..50b20c5 100644 --- a/docs/field_manuals/Field_Manual_06_Data_Poisoning_Playbook.md +++ b/docs/field_manuals/Field_Manual_06_Data_Poisoning_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 06: Data Poisoning -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Training Data / RAG Document Poisoning **Difficulty**: ⭐⭐⭐ Advanced diff --git a/docs/field_manuals/Field_Manual_07_Model_Theft_Playbook.md b/docs/field_manuals/Field_Manual_07_Model_Theft_Playbook.md index 573b273..378346d 100644 --- a/docs/field_manuals/Field_Manual_07_Model_Theft_Playbook.md +++ b/docs/field_manuals/Field_Manual_07_Model_Theft_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 07: Model Theft & Extraction -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Model Stealing / Extraction **Difficulty**: ⭐⭐⭐ Advanced diff --git a/docs/field_manuals/Field_Manual_08_DoS_Playbook.md b/docs/field_manuals/Field_Manual_08_DoS_Playbook.md index fd8e260..7328655 100644 --- a/docs/field_manuals/Field_Manual_08_DoS_Playbook.md +++ b/docs/field_manuals/Field_Manual_08_DoS_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 08: DoS & Resource Exhaustion -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Denial of Service & Cost Inflation **Difficulty**: ⭐⭐ Intermediate diff --git a/docs/field_manuals/Field_Manual_09_Multimodal_Playbook.md b/docs/field_manuals/Field_Manual_09_Multimodal_Playbook.md index 1b2a37c..9318b0c 100644 --- a/docs/field_manuals/Field_Manual_09_Multimodal_Playbook.md +++ b/docs/field_manuals/Field_Manual_09_Multimodal_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 09: Multimodal Attacks -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Vision/Audio + LLM Exploitation **Difficulty**: ⭐⭐ Intermediate diff --git a/docs/field_manuals/Field_Manual_10_Persistence_Playbook.md b/docs/field_manuals/Field_Manual_10_Persistence_Playbook.md index 4d0f7c8..f28ffcd 100644 --- a/docs/field_manuals/Field_Manual_10_Persistence_Playbook.md +++ b/docs/field_manuals/Field_Manual_10_Persistence_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 10: Persistence & Chaining -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: Multi-Turn Attack Sequences **Difficulty**: ⭐⭐⭐ Advanced diff --git a/docs/field_manuals/Field_Manual_11_Social_Engineering_Playbook.md b/docs/field_manuals/Field_Manual_11_Social_Engineering_Playbook.md index e7da8d4..1ca378b 100644 --- a/docs/field_manuals/Field_Manual_11_Social_Engineering_Playbook.md +++ b/docs/field_manuals/Field_Manual_11_Social_Engineering_Playbook.md @@ -1,6 +1,6 @@ # Field Manual Playbook 11: Social Engineering with LLMs -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) **Attack Type**: AI-Powered Phishing & Impersonation **Difficulty**: ⭐⭐ Intermediate diff --git a/docs/field_manuals/Field_Manual_Quick_Reference.md b/docs/field_manuals/Field_Manual_Quick_Reference.md index dfe8e6f..290e398 100644 --- a/docs/field_manuals/Field_Manual_Quick_Reference.md +++ b/docs/field_manuals/Field_Manual_Quick_Reference.md @@ -1,6 +1,6 @@ # Field Manual - Quick Reference Card -![ ](../assets/page_header.svg) +![ ](../assets/page_header_half_height.png) ## One-page cheat sheet for junior testers diff --git a/docs/templates/Chapter_Template.md b/docs/templates/Chapter_Template.md index b948771..51dd44a 100644 --- a/docs/templates/Chapter_Template.md +++ b/docs/templates/Chapter_Template.md @@ -11,7 +11,7 @@ Related: Chapters [list related chapters] # Chapter [X]: [Chapter Title] -![ ](assets/page_header.svg) +![ ](assets/page_header_half_height.png) _[Write a compelling 2-3 sentence abstract that: (1) describes what this chapter covers, (2) explains why it matters for AI red teaming, and (3) includes specific techniques/concepts covered. Example: "This chapter provides comprehensive coverage of [topic], including [technique 1], [technique 2], [technique 3], detection methods, defense strategies, and critical ethical considerations." Be specific and engaging.]_