aiml-moderation-content

What problem does it solve?

This Skill provides a structured, end-to-end benchmark for evaluating content moderation across multiple data variants, helping teams measure the effectiveness of safety filters in LLM workflows.

Core Features & Use Cases

• Multi-variant evaluation: covers prompt content, user prompts, and model outputs to test recall and robustness.
• Self-contained scoring: uses a lightweight rubric to surface strengths and gaps in model safety without relying on external services.
• Use Case: QA teams can run the benchmark to compare safety performance across models or updates, ensuring consistent guardrails before deployment.

Quick Start

Run the aiml-moderation-content benchmark with the included prompts to evaluate content moderation performance across all three variants.