Your AI assistant, now with a security and SEO reviewer
The Wardzo MCP server (coming soon) will let Claude, Cursor, ChatGPT and other MCP clients scan a site, read your latest report and explain each finding right inside your editor.
Status: coming soon. The MCP server is not available yet. Everything below describes what we are building. Until then, every report has a Copy as Markdown button so you can paste findings into any assistant.
If you build with AI, that is also where you fix things. The Wardzo MCP server will bring your monitoring data into the conversation, so your assistant can check a deploy and fix what it finds without you switching tabs.
Planned tools
scan_site: run a passive scan of a URL and return the score and findings.get_latest_report: fetch the most recent report for one of your monitored sites.explain_finding: return the why and how-to-fix for a finding, with framework-specific hints.list_sites: list your monitored sites and their current scores.
Planned limits by plan
Free: 20 calls per day. Basic: 500 calls per month. Premium: 5,000 calls per month. See pricing.
Frequently asked questions
Is the MCP server available now?
Not yet. It is in development and will be available to every plan at launch, including Free with a daily call limit.
What is MCP?
The Model Context Protocol is an open standard that lets AI assistants call external tools. Clients such as Claude, Cursor and ChatGPT can connect to MCP servers.
Related
- MCP setup for Claude, Cursor and ChatGPT
How the Wardzo MCP server will connect to Claude, Cursor and ChatGPT. The MCP server is coming soon; this page explains the planned setup.
- Website security monitoring
Outside-in security monitoring for live websites. Wardzo finds leaked API keys in JavaScript, exposed .env and .git files, weak security headers, SSL problems and cloaking, then tells you how to fix them.
- SEO health monitoring
Catch the SEO mistakes that silently remove you from Google and AI search, including accidental noindex, robots.txt blocks, missing sitemaps, wrong canonicals, broken pages and blocked AI crawlers.