LintBase MCP server

LintBase is a helper that looks at your live Firestore database and tells your AI what is actually inside it. It is handy for developers who use AI coding tools like Cursor, Claude Desktop, or Windsurf and keep getting code that references fields that do not exist. If your database has drifted away from your code, this is for you.

What this MCP server does

You ask your AI something like "add a field to users". The AI calls this LintBase helper. The helper looks at your live Firestore collections and reports the real field names, types, and how often each field appears. The AI then writes code based on what is actually there. You get a query or a change that matches your real data instead of a made-up one.

Click to zoom

What you can do with it

• Check the real field names in a Firestore collection before writing code
• See what types a field actually has across your documents
• Spot fields that appear in most documents but are missing in some
• Find collections that are readable without authentication
• Notice fields that look like personal data, such as emails or phone numbers
• See which collections are large enough to cost real money to read
• Get a summary of your schema so your AI stops inventing fields

Try asking your AI

• “What fields does my users collection actually have?”
• “Add an isVerified field to the users collection and show me the code”
• “Which collections in my Firestore have no authentication check?”
• “Is the price field in products stored as a number or a string?”

What it gives back to you

It gives your AI a summary of your Firestore schema: collection names, real field names, types, and how often each field appears. It can also report issues like missing auth rules or fields with mixed types. Your AI then uses that to answer you or write code. You see the result as a normal reply in your chat, not as a raw file.

Before you start

Who it's for

Developers and small teams who use AI coding tools and work with Firestore or MongoDB databases.