preetideepaksoni/AI-Security-Simulator — reverse-engineered prompt

Build me a local browser app called ASTRA that helps people learn AI security safely. It should run on my machine with Python and open in the browser, with a clean dark themed interface and simple sidebar navigation.

I want users to learn, attack, defend, respond, and practice. Include lessons and quizzes for AI security basics, prompt injection, data poisoning, adversarial examples, model extraction, jailbreaking, privacy attacks, supply chain risks, incident response, and governance. Add a vulnerable chatbot lab where people can try prompt injection only against the built in sandbox target, plus an attack catalog, automated red team report, defense lab with toggleable protections, token visualizer, forensics lab, incident response playbooks, and CTF style challenges with progress and badges.

Make it educational and ethical, with clear warnings not to test real systems. Include helpful sample content, reports, logs, and references to OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF, and EU AI Act. Look up current docs online if you need to.