Can I integrate these tests into my GitHub Actions?

Yes, the skill includes pre-configured YAML workflows for GitHub Actions to automate security scans, jailbreak tests, and safety evaluations on every push or PR.

How does the security gate feature work?

The security gate allows you to define failure thresholds for specific metrics, such as a zero-tolerance policy for critical vulnerabilities or specific jailbreak success rates, to block unsafe deployments.

What security frameworks does this skill support?

The security-testing skill maps directly to the OWASP Top 10 for LLMs (2025) and the NIST AI Risk Management Framework (RMF) for measurement and management.

What specific AI vulnerabilities can it detect?

It tests for direct and indirect prompt injection, jailbreak attempts, PII/training data leakage, system prompt disclosure, and toxic content generation.

AI Security Testing Automation

Name: AI Security Testing Automation
Author: pluginagentmarketplace

bypluginagentmarketplace

•

Security & Testing

Automates comprehensive security validation and red teaming for AI/ML systems through continuous testing and CI/CD integration.

About

This skill provides a robust framework for securing AI models and pipelines against evolving threats like prompt injection, jailbreaks, and data leakage. It automates complex security audits by mapping results to the OWASP LLM Top 10 and NIST AI RMF, enabling developers to implement rigorous security gates within their CI/CD pipelines. By providing actionable metrics and automated vulnerability assessments, it ensures AI applications remain safe, compliant, and resilient throughout the development lifecycle.

Key Features

1 GitHub stars
Continuous data leakage and PII disclosure detection
Full OWASP LLM Top 10 and NIST AI RMF coverage
Integrated CI/CD security gates with customizable thresholds
Automated prompt injection and jailbreak resistance testing
Detailed security reporting and vulnerability severity classification

Use Cases

Automating daily security audits for AI-powered APIs and infrastructure
Enforcing security compliance standards within automated DevOps pipelines
Validating LLM resilience against adversarial attacks before production deployment

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add pluginagentmarketplace/custom-plugin-ai-red-teaming security-testing

For use in Claude.ai and ChatGPT

Download Skill

GitHub