credibility-index/WebSec-AI ? reverse-engineered prompt

Build me a Python app called WebSecAI that scans a website for common security problems like SQL injection, XSS, CSRF, and SSRF, then gives me a clear report I can actually understand. I want both a simple web interface and a command line mode, plus a small REST API so I can kick off scans and check results later.

The app should let me enter a URL, run different scan profiles for quick checks or deeper testing, and optionally look for flags and other signs of exposed secrets like robots.txt, .git, .env, cookies, and hidden files. It should also include a recon mode for checking useful public info about a domain.

When the scan finishes, generate a markdown report in English and Russian with severity, impact, and practical fix advice. If AI features are enabled with an OpenRouter key, use them to explain findings and give remediation tips, but if the key is missing, show a clear message instead of pretending it works. Please make it easy to run locally and keep the UI clean and simple.