• DISCOVERY
  • QUERYING
  • VALIDATION
  • HUMAN

How to map API schemas with your AI agent

Connect your remote APIs once. Your agent then builds its own map of the data structures it needs to query.

Ask AI about this page

Short answer

How can I map API schemas with my AI?

The Connector scans your remote endpoints to build a structured map of tables and columns. Your agent uses this map to understand exactly what data is available and how to query it. You get a clear view of the remote data structure.

Where your data maps land

The schema mapping outcomes

The AI turns raw API endpoints into a structured map your agent can actually use.

DISCOVERY

Unified data maps

The AI identifies all available endpoints and organizes them into a clean, searchable schema. You get a clear view of the remote data structure.

QUERYING

Ready to query

Once the schema is mapped, your agent knows which columns to pull. You can ask questions about specific data points immediately.

VALIDATION

Schema accuracy

The AI checks column types and table names against the live API. This ensures your agent doesn't try to query fields that don't exist.

HUMAN

Manual overrides

If an API uses cryptic field names, you decide how to rename them. The AI follows your lead to make the schema readable.

The workflow

What your AI does when the connection is live.

The AI moves from finding the connection to understanding the specific data points within it.

  1. Find connections

    The AI looks through your active accounts to see which remote services are ready to be mapped.

    cdata_list_connections

  2. Inspect metadata

    The agent pulls the high-level structure of the API to understand the available data objects.

    cdata_get_schema_metadata

  3. List tables

    The AI identifies the specific tables or resource collections available within the API structure.

    cdata_list_tables

  4. Map columns

    The agent drills down into each table to find the exact field names and data types.

    cdata_get_table_columns

Try it

Copy these to start.

Use these prompts to begin mapping your data.

Starting points

Replace the bracketed text with your specific connection names or table titles.

CData Connect Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_SBYp4YpOhOkod1f5osCBPGeYZMDXyp0rzOXZLCrf/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 — CData Connect capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "cdata-connect-cloud-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_SBYp4YpOhOkod1f5osCBPGeYZMDXyp0rzOXZLCrf/mcp"
    }
  }
}
Copy into chat04

  • List all my available connections and show me the schema for the Salesforce connection.

  • What tables are available in my HubSpot connection?

  • Show me the columns for the 'Orders' table in my Shopify API.

  • Map the schema for my Zendesk connection so I can start querying tickets.

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

Start here

Connect CData Connect Cloud once, then ask.

Just link your account once. Your credentials stay encrypted and managed by Vinkius, and you can immediately start asking your agent to explore your data.

Connect CData Connect Cloud to your AI

FAQ

How the mapping behaves

  • 01

    Can the AI change data in my remote API?

    No. These tools are for discovering and reading schema metadata. The AI cannot modify, delete, or create data in your remote system.

  • 02

    Does the AI automatically rename my API fields?

    No. The AI maps the existing names. You must manually instruct the agent if you want it to use different aliases for specific fields.

  • 03

    How does the AI know which columns are available?

    The AI uses the schema metadata tools to scan the remote endpoint and build a list of every available table and column.

  • 04

    Can I use this with any AI client?

    Yes, as long as your client is MCP-compatible, such as Claude, Cursor, or Windsurf.

  • 05

    Will the AI see my API passwords?

    No. Your credentials are encrypted and managed by Vinkius. The AI only interacts with the data structure provided through the tools.

  • More questions about CData Connect Cloud? The Connector page answers them. See everything the CData Connect Cloud Connector can do

Connect CData Connect to Claude, Cursor, ChatGPT & more