Version updated for https://github.com/arpitbhasin1/ai-security-ci to version v1.0.1. This action is used across all versions by ? repositories. Go to the GitHub Marketplace to find the latest changes. Action Summary The AI Security CI GitHub Action automates security testing for AI systems within CI pipelines by simulating prompt-based attacks to identify vulnerabilities, such as jailbreak attempts, prompt leakage, and harmful content generation. It evaluates responses using heuristics and an optional LLM-based judge, providing detailed JSON and Markdown reports.

Version updated for https://github.com/arpitbhasin1/ai-security-ci to version v1. This action is used across all versions by ? repositories. Go to the GitHub Marketplace to find the latest changes. Action Summary This GitHub Action automates security testing for AI systems by simulating common prompt-based attacks, such as jailbreak attempts, prompt leakage, and harmful content generation. It identifies vulnerabilities by evaluating system responses using heuristics and optional LLM-based judging, providing detailed reports in JSON and Markdown formats.