MCP Gateway Scan server

This is a small helper that checks a folder of code and tells you if it looks ready for real use. It looks for common mistakes like secrets written in plain text, missing limits on spending, and error handling that quietly lets things through. It is handy for developers and anyone who reviews code before it goes live.

What this MCP server does

You ask your AI to scan a folder or a file. The AI passes that request to this helper. The helper reads the files in that folder, looking for patterns that often cause problems in real systems. It then gives back a simple red, yellow, or green score across seven areas, plus the exact lines where it found issues. It never runs your code, never goes online, and never shows secret values, only where they are.

Click to zoom

What you can do with it

• Scan a folder of code and get a safety score
• Find places where secrets are written in plain text
• Check if error handling fails open instead of closed
• Spot missing spending limits or rate limits
• See if logging and tracing are set up
• Get a connect or do-not-connect verdict for a third-party server
• Run the scan in a pipeline and fail the build on red issues

Try asking your AI

• “Scan the folder ./my-gateway and tell me if it is ready for production”
• “Check ./some-mcp-server and give me a connect verdict”
• “Run a gateway readiness scan on ./gateway with the CI flag on”
• “Look at ./.mcp.json and tell me if it is safe to connect”

What it gives back to you

You get a short report in the chat with a score for each of seven areas, shown as green, yellow, or red. For each problem, it shows the file and line number where it found it. If you ask for JSON, you get the same information in a structured format. It never shows the actual secret values, only where they are.

Before you start

Who it's for

Developers, security reviewers, and anyone who checks code before it goes live.