LLM (Large Language Models)

# Prompting + Defensive Measures

Filtering Techniques: Blocklists and Allowlists for Safe AI Promptslearnprompting.org

# Types of Prompt Injections

Understanding Direct and Indirect AI Prompt Injections and Their ImplicationsEmbrace The Red

1. Direct Prompt Injections

2. Second Order Prompt Injections (aka Indirect Prompt Injections)

3. Cross-Context AI Injections

# Copied Prompt Injection PoC

ChatGPT PoCprompt-injection.onrender.com

# Insecure Response Processing [Data Exfiltration]

Do not blindly trust LLM responses. Threats to chatbots.Embrace The Red

# AI hallucinations

Cybersecurity Snapshot: New Guide Details How To Use AI Securely, as CERT Honcho Tells CISOs To Sharpen AI Security Skills ProntoTenable®

# Testing Frameworks [To-Do]

1. Giskard

GitHub - Giskard-AI/giskard-oss: 🐢 Open-Source Evaluation & Testing library for LLM AgentsGitHub

1. langflow

GitHub - langflow-ai/langflow: Langflow is a powerful tool for building and deploying AI-powered agents and workflows.GitHub

# Jailbreaking Chat/ Do Anything Now (DAN)

https://www.jailbreakchat.com/www.jailbreakchat.com

# Threat Modelling

https://aivillage.org/large%20language%20models/threat-modeling-llm/aivillage.org