Quickstart

This guide walks you through connecting a database, creating a tool, and exposing it to Claude — in under 5 minutes.

Prerequisites

• A Datatape account (sign up free)
• A database you want to query (PostgreSQL, MySQL, Snowflake, BigQuery, or Google Sheets)
• An AI client that supports MCP (Claude Desktop, Claude Code, Cursor, or ChatGPT)

Step 1: Create an organization

After signing in, you'll be prompted to create an organization. This is your workspace — all sources, tools, and team members live here.

Step 2: Connect a source

Go to Sources and click Add source.

1. Choose your database type (e.g., PostgreSQL)
2. Enter your connection details:
   • Host, port, database name
   • Username and password (read-only credentials recommended)
3. Click Test connection to verify
4. Click Save

Step 3: Create a tool

Go to Tools and click Create tool.

1. Give it a name and description (AI agents use the description to decide when to call it)
2. Select the source you just connected
3. Write a SQL template:

SELECT customer_name, total_spent
FROM customers
WHERE region = {{ region }}
ORDER BY total_spent DESC
LIMIT {{ limit }}

4. Add parameters:
   • region — type: string, required: true
   • limit — type: number, required: false, default: 10
5. Click Test to run the query with sample values
6. Click Save

Step 4: Create an agent

Go to Agents and click Create agent.

1. Give it a name and optional description (e.g., "Customer Data Bot")
2. Assign tools directly or via a toolset
3. Click Create

Your agent is created with an embedded API key. The full key is shown once — copy it and store it securely.

Step 5: Connect Claude

Your organization's MCP endpoint is:

https://your-org-id.mcp.datatape.ai/mcp

Claude Desktop

Add this to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "datatape": {
      "url": "https://your-org-id.mcp.datatape.ai/mcp"
    }
  }
}

Claude Code

claude mcp add datatape https://your-org-id.mcp.datatape.ai/mcp

Cursor

Go to Settings > MCP Servers > Add Server and paste your endpoint URL.

What's next?

• Connect more sources — PostgreSQL, Snowflake, BigQuery, and more
• Create advanced tools — Conditional clauses, Jinja2 templates
• Add team members — Invite your team and assign roles
• Create agents — Set up programmatic access with agent API keys