safety-evaluation

Community

Evaluate AI model safety across categories with rubrics and thresholds.

Authorjassics
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill addresses the need for comprehensive safety evaluation of AI models, providing a structured framework to measure and enhance safety across various harm categories.

Core Features & Use Cases

  • Structured Evaluation: Conducts safety evaluations structured across harm categories such as disallowed content, robustness, and truthfulness.
  • Rubrics and Thresholds: Implements clear rubrics and pass/fail thresholds for evaluating AI system performance.
  • Reporting: Generates detailed safety evaluation reports to inform release decisions and track regression over time.

Quick Start

Run the safety-evaluation skill with the provided prompts and evaluate the AI system's response against established harm categories.

Dependency Matrix

Required Modules

None required

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: safety-evaluation
Download link: https://github.com/jassics/awesome-claude-security/archive/main.zip#safety-evaluation

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.