mirror of
https://github.com/Shiva108/ai-llm-red-team-handbook.git
synced 2026-10-13 19:21:05 +02:00
docs: remove alt text from page header images in all chapters and field manuals
This commit is contained in:
shiva108 committed
1 parent
c8d1f38ffc
commit
4ca1780bc5
61 files changed
+138
-138
No files matched your search
@@ -1,6 +1,6 @@
|
||||
# Red Teaming AI & LLMs: The Consultant’s Complete Handbook
|
||||
|
||||
|
||||
|
||||
|
||||
## Table of Contents
|
||||
|
||||
@@ -104,7 +104,7 @@ _(To be added)_
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 1: Introduction to AI Red Teaming
|
||||
|
||||
@@ -187,7 +187,7 @@ The handbook is organized for practical learning and use:
|
||||
|
||||
_Proceed to the next chapter to explore ethical and legal essentials, and begin developing the professional approach required of every AI red teamer._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 2: Ethics, Legal, and Stakeholder Communication
|
||||
|
||||
@@ -289,7 +289,7 @@ In AI red teaming, technical findings may have legal, business, or even social i
|
||||
|
||||
_In the next chapter, you’ll develop the mindset that distinguishes effective AI red teamers from traditional security testers, bridging technology, psychology, and business acuity._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 3: The Red Teamer's Mindset
|
||||
|
||||
@@ -369,7 +369,7 @@ Field engagements can be high-stress: production outages, tense clients, critica
|
||||
|
||||
_Mastering the red team mindset primes you for the work ahead: scoping, planning, and then executing engagements with insight, rigor, and integrity. Proceed to the next chapter to learn how to prepare and manage a professional AI red team project from start to finish._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 4: SOW, Rules of Engagement, and Client Onboarding
|
||||
|
||||
@@ -488,7 +488,7 @@ Before you start:
|
||||
|
||||
_Solid foundations prevent project failure and foster trust. The next chapter will guide you through threat modeling and risk analysis for AI systems, helping you identify what matters most before you begin attacking._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 5: Threat Modeling and Risk Analysis
|
||||
|
||||
@@ -613,7 +613,7 @@ A good threat model is:
|
||||
|
||||
_With a strong threat model, your red team engagement becomes risk-driven and results-focused. The next chapter will walk you through scoping these findings into a feasible, valuable engagement plan._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 6: Scoping an Engagement
|
||||
|
||||
@@ -721,7 +721,7 @@ An accurately scoped engagement shows professionalism and respect for the client
|
||||
|
||||
_With a precise scope in place, you are ready to establish the laboratory, test environments, and safety measures needed for executing a secure and efficient AI red teaming exercise. Continue to the next chapter for practical lab setup and environmental safety._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 7: Lab Setup and Environmental Safety
|
||||
|
||||
@@ -816,7 +816,7 @@ Remember:
|
||||
|
||||
_With a robust lab and clear safety controls in place, you’re prepared to gather and preserve evidence in a trustworthy manner. Continue to the next chapter to master documentation and evidence handling in AI red team engagements._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 8: Evidence, Documentation, and Chain of Custody
|
||||
|
||||
@@ -930,9 +930,9 @@ A robust chain of custody ensures that all evidence remains trustworthy and trac
|
||||
|
||||
_With evidence and documentation in place, you’re equipped to deliver clear, credible findings. The next chapter will guide you through the art of writing actionable, impactful red team reports for both technical and executive audiences._
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 9: LLM Architectures and System Components
|
||||
|
||||
@@ -1071,11 +1071,11 @@ Before attacking, answer these questions about your target:
|
||||
|
||||
Understanding these components transitions you from "guessing passwords" to "engineering exploits."
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 10: Tokenization, Context, and Generation
|
||||
|
||||
|
||||
|
||||
|
||||
While the "mind" of an LLM is a neural network, its "senses" are defined by the Tokenizer, and its "memory" is defined by the Context Window. As a Red Teamer, deeply understanding these mechanisms allows you to exploit blind spots, bypass filters, and degrade model performance.
|
||||
|
||||
@@ -1288,7 +1288,7 @@ Understanding plugins is critical because they turn a "text generator" into an "
|
||||
|
||||
# Chapter 12: Retrieval-Augmented Generation (RAG) Pipelines
|
||||
|
||||
|
||||
|
||||
|
||||
## 12.1 What Is Retrieval-Augmented Generation (RAG)?
|
||||
|
||||
@@ -3815,7 +3815,7 @@ class RAGAccessControlTester:
|
||||
|
||||
_RAG systems represent one of the most powerful - and vulnerable - implementations of LLM technology in enterprise environments. By understanding their architecture, attack surfaces, and testing methodologies, red teamers can help organizations build secure, production-ready AI assistants. The next chapter will explore data provenance and supply chain security - critical for understanding where your AI system's data comes from and how it can be compromised._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 13: Data Provenance and Supply Chain Security
|
||||
|
||||
@@ -5676,7 +5676,7 @@ def detect_insider_poisoning(training_data, baseline_distribution):
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 14: Prompt Injection (Direct/Indirect, 1st/3rd Party)
|
||||
|
||||
@@ -9746,7 +9746,7 @@ _Prompt injection represents the defining security challenge of the LLM era. Lik
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 15: Data Leakage and Extraction
|
||||
|
||||
@@ -13416,7 +13416,7 @@ _Continue to Chapter 16: Jailbreaks and Bypass Techniques to learn how attackers
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 16: Jailbreaks and Bypass Techniques
|
||||
|
||||
@@ -14962,7 +14962,7 @@ Successful reports are tailored to multiple audiences, such as:
|
||||
|
||||
_You are now ready to communicate your findings with clarity and impact. The next chapter will cover presenting results to both technical and non-technical stakeholders - ensuring your work leads to measurable improvements in AI security._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 37: Presenting Results and Remediation Guidance
|
||||
|
||||
@@ -15048,7 +15048,7 @@ Delivering findings is more than handing over a report - it's about ensuring you
|
||||
|
||||
_Professional communication and practical remediation guidance ensure your red teaming work translates into real, measurable improvements. The next chapter will explore lessons learned, common pitfalls, and how to build a mature AI/LLM red teaming practice._
|
||||
|
||||
|
||||
|
||||
|
||||
# Chapter 38: Lessons Learned and Building Future Readiness
|
||||
|
||||
@@ -15108,4 +15108,4 @@ To make AI red teaming a sustainable part of your organization’s security post
|
||||
|
||||
_By systematically learning and adapting, your AI red teaming program matures - helping organizations stay resilient amid the evolving risks and rewards of intelligent systems._
|
||||
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# **AI/LLM Red Team Field Manual**
|
||||
|
||||
|
||||
|
||||
> **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
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **1. Introduction: Rules of Engagement (RoE)**
|
||||
|
||||
@@ -204,7 +204,7 @@ Define in writing: in-scope systems/models, allowed techniques, test windows, ha
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **1.5 Environment Setup & Configuration**
|
||||
|
||||
@@ -490,7 +490,7 @@ API Config File: ✅
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **2. Red Teaming Phases**
|
||||
|
||||
@@ -765,7 +765,7 @@ chmod +x cleanup.sh
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **3\. Attack Types & Practical Test Examples**
|
||||
|
||||
@@ -1250,7 +1250,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **4\. Tools Reference & CLI Commands**
|
||||
|
||||
@@ -1296,7 +1296,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **5\. Attack-Type–to–Tool Quick Lookup Table**
|
||||
|
||||
@@ -1315,7 +1315,7 @@ afl-fuzz \-i testcase_dir \-o findings_dir \-- ./your_cli_target @@
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **5.5 API Configuration Guide**
|
||||
|
||||
@@ -1583,7 +1583,7 @@ echo "✅ Provider test complete!"
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **6\. Reporting Guidance**
|
||||
|
||||
@@ -1597,7 +1597,7 @@ Report every finding with:
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **7\. Additional Guidance**
|
||||
|
||||
@@ -1609,7 +1609,7 @@ Report every finding with:
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
|
||||
## **8. Troubleshooting Guide**
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 1: Introduction to AI Red Teaming
|
||||
|
||||
|
||||
|
||||
|
||||
## 1.1 What Is AI Red Teaming?
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 2: Ethics, Legal, and Stakeholder Communication
|
||||
|
||||
|
||||
|
||||
|
||||
## 2.1 Why Ethics Matter in AI Red Teaming
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 3: The Red Teamer's Mindset
|
||||
|
||||
|
||||
|
||||
## 3.1 What Sets a Red Teamer Apart?
|
||||
|
||||
Unlike traditional vulnerability assessors or automated scanning, a red teamer adopts the mindset of a determined, creative, and unpredictable adversary. Great red teamers aren’t just tool users: they are critical thinkers, problem solvers, and empathetic adversaries who model real-world threats with nuance and rigor.
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 4: SOW, Rules of Engagement, and Client Onboarding
|
||||
|
||||
|
||||
|
||||
## 4.1 The Foundation of a Secure Engagement
|
||||
|
||||
Before any AI red teaming begins, you must have clearly agreed-upon definitions of what, how, and when you are allowed to test. This is formalized through three key processes:
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 5: Threat Modeling and Risk Analysis
|
||||
|
||||
|
||||
|
||||
## 5.1 Why Threat Modeling Matters in AI Red Teaming
|
||||
|
||||
Threat modeling is a proactive process that helps you and stakeholders understand **what’s at risk, who might attack, and how they could succeed**. In AI/LLM systems, the landscape is especially dynamic: you must account for unique risks like model manipulation, data leakage via prompts, unintended plugin behavior, and more.
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 6: Scoping an Engagement
|
||||
|
||||
|
||||
|
||||
## 6.1 The Importance of Proper Scoping
|
||||
|
||||
A well-scoped engagement ensures that the red teaming exercise is effective, safe, focused, and delivers value to the client. Poor scoping can lead to missed risks, out-of-control timelines, client confusion, or legal exposure. In AI red teaming, scoping must adapt to the unique complexities and dynamic nature of machine learning systems, APIs, plugins, and data flows.
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 7: Lab Setup and Environmental Safety
|
||||
|
||||
|
||||
|
||||
## 7.1 Why Lab Setup and Environmental Safety Matter
|
||||
|
||||
A properly designed test environment (or "lab") is crucial in AI red teaming to:
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 8: Evidence, Documentation, and Chain of Custody
|
||||
|
||||
|
||||
|
||||
## 8.1 The Role of Evidence in Red Teaming
|
||||
|
||||
Evidence is the backbone of credible red team engagements. In AI/LLM systems, good evidence ensures that:
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 9: LLM Architectures and System Components
|
||||
|
||||
|
||||
|
||||
Effective Red Teaming requires moving beyond treating AI as a "black box." To identify subtle vulnerabilities, bypass guardrails, or exploit system-level integration flaws, you must understand the underlying architecture. This chapter deconstructs Large Language Models (LLMs) and their ecosystem from an adversarial perspective.
|
||||
|
||||
## 9.1 The AI Attack Surface
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 10: Tokenization, Context, and Generation
|
||||
|
||||
|
||||
|
||||
|
||||
While the "mind" of an LLM is a neural network, its "senses" are defined by the Tokenizer, and its "memory" is defined by the Context Window. As a Red Teamer, deeply understanding these mechanisms allows you to exploit blind spots, bypass filters, and degrade model performance.
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 11: Plugins, Extensions, and External APIs
|
||||
|
||||
|
||||
|
||||
|
||||
Modern LLMs are no longer isolated "chatbots." Through plugins, functions, and extensions, they can browse the web, read emails, query databases, and execute code. This capability introduces the **Tool-Use Attack Surface**, where the LLM becomes a "privileged user" that attackers can manipulate.
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 12: Retrieval-Augmented Generation (RAG) Pipelines
|
||||
|
||||
|
||||
|
||||
|
||||
## 12.1 What Is Retrieval-Augmented Generation (RAG)?
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 13: Data Provenance and Supply Chain Security
|
||||
|
||||
|
||||
|
||||
|
||||
## 13.1 Understanding Data Provenance in AI/LLM Systems
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 14: Prompt Injection (Direct/Indirect, 1st/3rd Party)
|
||||
|
||||
|
||||
|
||||
|
||||
## 14.1 Introduction to Prompt Injection
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 15: Data Leakage and Extraction
|
||||
|
||||
|
||||
|
||||
|
||||
## 15.1 Introduction to Data Leakage in LLMs
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 16: Jailbreaks and Bypass Techniques
|
||||
|
||||
|
||||
|
||||
_This chapter provides comprehensive coverage of jailbreak techniques, bypass methods, testing methodologies, and defenses for LLM systems._
|
||||
|
||||
## 16.1 Introduction to Jailbreaking
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 17: Plugin and API Exploitation
|
||||
|
||||
|
||||
|
||||
_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._
|
||||
|
||||
## 17.1 Introduction to Plugin and API Security
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# 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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# 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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# 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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 21: Model DoS and Resource Exhaustion
|
||||
|
||||
|
||||
|
||||
_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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# 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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# 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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 24: Social Engineering with LLM
|
||||
|
||||
|
||||
|
||||
_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._
|
||||
|
||||
## Introduction
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 25: Advanced Adversarial ML
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 26: Supply Chain Attacks on AI
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 27: Federated Learning Attacks
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 28: AI Privacy Attacks
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 29: Model Inversion Attacks
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 30: Backdoor Attacks
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 31: AI System Reconnaissance
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 32: Automated Attack Frameworks
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 33: Red Team Automation
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 34: Defense Evasion Techniques
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 35: Post-Exploitation in AI Systems
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 36: Reporting and Communication
|
||||
|
||||
|
||||
|
||||
|
||||
## 36.1 The Purpose of Red Team Reports
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 37: Remediation Strategies
|
||||
|
||||
|
||||
|
||||
|
||||
## 37.1 The Importance of Presentation
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 38: Continuous Red Teaming
|
||||
|
||||
|
||||
|
||||
|
||||
## 38.1 Common Pitfalls in AI/LLM Red Teaming
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 39: AI Bug Bounty Programs
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 40: Compliance and Standards
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 41: Industry Best Practices
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 42: Case Studies and War Stories
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 43: Future of AI Red Teaming
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 44: Emerging Threats
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Chapter 45: Building an AI Red Team Program
|
||||
|
||||
|
||||
|
||||
|
||||
## 1.0 Introduction: The Imperative for AI-Specific Red Teaming
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Chapter 46: Conclusion and Next Steps
|
||||
|
||||
|
||||
|
||||
_This chapter is currently under development._
|
||||
|
||||
## TBD
|
||||
|
||||
+1
-1
@@ -2,7 +2,7 @@
|
||||
|
||||
Version 0.024.63
|
||||
|
||||
|
||||
|
||||
|
||||
## Introduction
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 01: Prompt Injection
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Prompt Injection (Direct & Indirect)
|
||||
**Difficulty**: ⭐ Beginner
|
||||
**OWASP LLM**: #1 | **MITRE ATLAS**: T0803
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 02: Data Leakage & Extraction
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Training Data & Memory Extraction
|
||||
**Difficulty**: ⭐⭐ Intermediate
|
||||
**OWASP LLM**: #6 | **MITRE ATLAS**: T0802
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 03: Jailbreaks & Bypass
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Content Filter & Safety Bypass
|
||||
**Difficulty**: ⭐ Beginner
|
||||
**OWASP LLM**: #1 | **MITRE ATLAS**: T0803
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 04: Plugin & API Exploitation
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Plugin/Function Calling Exploitation
|
||||
**Difficulty**: ⭐⭐⭐ Advanced
|
||||
**OWASP LLM**: #7 | **MITRE ATLAS**: T0806
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 05: Evasion & Obfuscation
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Input Filter Bypass
|
||||
**Difficulty**: ⭐⭐ Intermediate
|
||||
**OWASP LLM**: #1 | **Focus**: Filter Evasion
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 06: Data Poisoning
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Training Data / RAG Document Poisoning
|
||||
**Difficulty**: ⭐⭐⭐ Advanced
|
||||
**OWASP LLM**: #3 | **MITRE ATLAS**: T0808
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 07: Model Theft & Extraction
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Model Stealing / Extraction
|
||||
**Difficulty**: ⭐⭐⭐ Advanced
|
||||
**OWASP LLM**: #10 | **MITRE ATLAS**: T0809
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 08: DoS & Resource Exhaustion
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Denial of Service & Cost Inflation
|
||||
**Difficulty**: ⭐⭐ Intermediate
|
||||
**OWASP LLM**: #4 | **MITRE ATLAS**: T0804
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 09: Multimodal Attacks
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Vision/Audio + LLM Exploitation
|
||||
**Difficulty**: ⭐⭐ Intermediate
|
||||
**OWASP LLM**: #1 | **Focus**: Cross-Modal Injection
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 10: Persistence & Chaining
|
||||
|
||||
|
||||
|
||||
**Attack Type**: Multi-Turn Attack Sequences
|
||||
**Difficulty**: ⭐⭐⭐ Advanced
|
||||
**OWASP LLM**: #1 | **Focus**: Attack Chains
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual Playbook 11: Social Engineering with LLMs
|
||||
|
||||
|
||||
|
||||
**Attack Type**: AI-Powered Phishing & Impersonation
|
||||
**Difficulty**: ⭐⭐ Intermediate
|
||||
**OWASP LLM**: Multiple | **Ethical**: HIGH RISK
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
|
||||
# Field Manual - Quick Reference Card
|
||||
|
||||
|
||||
|
||||
**One-page cheat sheet for junior testers**
|
||||
|
||||
---
|
||||
|
||||
Reference in new issue
Block a user