GuardVibe MCP server

GuardVibe is a security scanner that plugs into the AI coding assistant you already use. It checks the code your AI writes for common security mistakes, like exposed passwords or outdated packages with known bugs. It is handy if you are building a web app with AI help and want a second pair of eyes on what gets produced.

What this MCP server does

You ask your AI to check your project for security issues. Your AI passes that request to GuardVibe, which reads your code files and dependency list. GuardVibe compares what it finds against hundreds of security rules and a daily-updated list of known vulnerable package versions. It then hands back a list of findings, a security score, and suggested fixes your AI can apply. You see the results in your chat window or in a report file.

Click to zoom

What you can do with it

• Scan your whole project for security problems and get a score
• Check your package list for known vulnerable versions
• Find hardcoded passwords and API keys in your code
• Spot missing login checks on your web routes
• Get concrete fix suggestions your AI can apply
• Check your AI tool setup for unsafe settings
• Run a security check before you commit code

Try asking your AI

• “Scan my project for security issues and tell me what to fix first”
• “Are any of my npm packages known to be vulnerable?”
• “Check if I left any API keys or passwords in my code”
• “Look at my Next.js routes and tell me which ones are missing login checks”

What it gives back to you

You get a list of findings grouped by type, each with a short explanation and a severity level. You also get a security score for your project and a checklist of fixes. If you ask for a fix, your AI can show you the exact code change to make. Everything appears right in your chat or in a report file you can open.

Before you start

Who it's for

Developers and makers who build web apps with AI coding assistants and want a simple security check without setting up a big security platform.