A sovereign AI constitution with nine Laws, falsification protocols, and an executable reference engine. Platform‑agnostic. Auditable. Adoptable.

LLM Red Team security testing for LLMs and LLM-driven applications, built for red teams and whitehats. Generate adversarial suites from security playbooks (OWASP LLM, OWASP Agent, MITRE ATLAS, jailbreak, multimodal/file-upload), execute them against live targets via browser UI or HTTP API, and assess risk.

🦄 4-This repository presents a concise framework for AI governance and compliance, aligned with the EU AI Act, Brazil’s PL 2338/2023, and ethical guidelines from the CFP and Stanford HAI, with a focus on high-risk EmotionAI applications in psychology and education.

An automated pipeline to audit video content for regulatory compliance using LangGraph, Azure Video Indexer, Azure AI Search, and Azure OpenAI (GPT-4o). It leverages LangSmith for optimization and Azure Application Insights for monitoring, generating structured JSON compliance reports.

AIRTA is an open source, production-ready AI Risk Testing Agent. Point it at your chatbot, copilot, or API; build structured compliance tests from rubrics such as the EU AI Act and OECD; then assess every response against regulatory mandates.