• SCHEMA
  • INDEXES
  • MAPPING
  • HUMAN

How to inspect database schema with your AI

Connect Fauna to your AI client once. From there, your agent can read your entire database structure to find mismatches or missing indexes.

Ask AI about this page

Short answer

Can my AI check my Fauna database structure?

Yes. Your AI uses the execute_fql tool to run queries that pull your collection definitions and index settings. This allows your agent to compare your actual schema against your app's expected data model.

Where your findings land.

What your AI finds

The AI scans your Fauna setup to identify structural gaps.

SCHEMA

Collection audits

The AI reads your collection definitions to confirm they hold the right fields. It catches missing attributes before they break your app.

INDEXES

Index verification

Your agent checks if your indexes support your most frequent queries. It identifies where missing indexes might cause slow performance.

MAPPING

Model alignment

The AI compares your database structure to your application code. It flags where your data types or nesting don't match your logic.

HUMAN

Manual fixes

The AI identifies the errors but doesn't change your database. You still handle the actual schema migrations and index creation.

The workflow

What your AI does when the request arrives.

Your agent acts as a database administrator by translating your questions into FQL commands.

  1. Fetch schema

    The AI pulls the current state of your collections and indexes from Fauna.

    execute_fql

  2. Analyze structure

    It parses the returned data to understand how your documents are organized.

    execute_fql

  3. Compare models

    The agent checks the retrieved schema against the requirements you provided in chat.

    execute_fql

  4. Report gaps

    The AI lists specific missing indexes or mismatched fields in a clear summary.

    execute_fql

Try it

Copy these to start.

Use these prompts to begin your audit.

Starting points

Replace the bracketed text with your specific collection or index names.

Fauna Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_hEG5bwdrY3FdPx62RCPAMcIqUBDzab1jcyYhjY8I/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Fauna capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "fauna-serverless-db-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_hEG5bwdrY3FdPx62RCPAMcIqUBDzab1jcyYhjY8I/mcp"
    }
  }
}
Copy into chat04

  • List all the collections in my Fauna database and show me their schemas.

  • Check if my 'users' collection has the necessary indexes to support queries by email.

  • Compare my current Fauna schema to this application model: [paste model code].

  • Find any collections that are missing the required 'createdAt' timestamp field.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Fauna once, then ask.

You only need to link your account once. Your credentials stay encrypted, and you can immediately start asking your AI about your database.

Connect Fauna to your AI

FAQ

How this task behaves

  • 01

    Can the AI change my database schema?

    No. The AI can only read your schema using execute_fql. It cannot create, delete, or modify collections or indexes.

  • 02

    Does the AI see all my data?

    The AI only sees the data it needs to answer your specific question. If you ask for a schema audit, it focuses on metadata rather than individual document contents.

  • 03

    How does the AI run queries?

    It uses the execute_fql tool to send commands directly to your Fauna database.

  • 04

    Can it fix my missing indexes?

    No. The AI identifies that an index is missing, but you must run the command to create it yourself.

  • 05

    Will this work with any AI client?

    Yes, it works with any client that supports the Model Context Protocol.

  • More questions about Fauna? See everything the Fauna Connector can do

Connect Fauna to Claude, Cursor, ChatGPT & more