LLM Testing Skill

LLM Testing Skill

Description

Comprehensive testing prompts and wordlists for evaluating Large Language Model (LLM) security, safety, and robustness. This skill provides curated test cases for bias detection, data leakage prevention, alignment testing, privacy boundaries, and adversarial prompt resistance.

Purpose

This skill is designed for security researchers, AI safety engineers, and red team operators to assess LLM vulnerabilities and ensure models:

• Maintain ethical boundaries and safety guidelines
• Avoid data leakage and privacy violations
• Resist bias and discrimination
• Prevent adversarial prompt attacks
• Maintain alignment with intended behavior

Authorized Use Cases

• AI Safety Research: Testing LLM models for safety and alignment issues
• Security Auditing: Red team operations for AI/ML systems
• Bias Detection: Evaluating models for fairness and bias
• Privacy Testing: Verifying models don't leak training data or PII
• Educational Purposes: Teaching AI security concepts and vulnerabilities
• Responsible AI Development: Building more robust and secure LLMs

Test Categories

1. Bias Testing

Test for biases in LLM responses across various dimensions:

Gender Bias (Bias_Testing/gender_bias.txt)

• Prompts to reveal gender stereotypes and biases
• Career and role associations
• Behavioral and capability assumptions
• Note: Replace [GENDER] placeholder before use

Nationality & Geographic Bias (Bias_Testing/nationality_geographic_bias.txt)

• Cultural stereotypes and geographic prejudices
• Regional capability assumptions
• Note: Replace [COUNTRY], [REGION], [NATIONALITY] placeholders