Documentation - Dolores MCP

What is this?

This is a No-Code MCP Server. MCP (Model Context Protocol) is a standard that lets AI assistants like Claude call external tools — things like adding data to a spreadsheet, looking up information, or triggering an action.

With this tool, you can create custom MCP tools by writing JavaScript code in your browser. No deployment or infrastructure needed — just write the code, test it, and connect it to your AI assistant.

Step 1: Create an Account

Go to /web/signup and create an account with your email, password, and company name.

Step 2: Create a Tool

Go to the Tools page
Click + Add Tool
Fill in:
- Tool Name — a unique identifier (letters, numbers, underscores only, e.g. add_student)
- Description — what the tool does (the AI reads this to decide when to use it)
- Input Fields — the data the AI will pass to your tool (e.g. name, email, phone)
Click Add Tool

Step 3: Write the Code

Each tool has a JavaScript code editor. This code runs when the AI calls your tool. You can also click AI Generate to have AI write the code for you.

Your code has access to:

args — the input from the AI (e.g. args.name, args.phone)
google — the Google APIs library (for Sheets, Drive, etc.)
console.log() — for debugging (output is captured)

Your code must return a string — this is sent back to the AI as the tool's result.

Example: Add a row to Google Sheets

const auth = new google.auth.GoogleAuth({
  credentials: {
    type: 'service_account',
    client_email: 'your-sa@project.iam.gserviceaccount.com',
    private_key: '-----BEGIN PRIVATE KEY-----\n...\n-----END PRIVATE KEY-----\n',
  },
  scopes: ['https://www.googleapis.com/auth/spreadsheets'],
});

const sheets = google.sheets({ version: 'v4', auth });
await sheets.spreadsheets.values.append({
  spreadsheetId: 'YOUR_SHEET_ID',
  range: 'Sheet1!A1',
  valueInputOption: 'USER_ENTERED',
  insertDataOption: 'INSERT_ROWS',
  requestBody: {
    values: [[args.name, args.phone, new Date().toISOString()]]
  },
});

return `Student ${args.name} recorded successfully.`;

Step 4: Test Your Tool

Click the Test button on your tool card. This saves your code and runs it with sample data (e.g. test_name, test_phone). Check the result to make sure it works before connecting it to your AI.

Step 5: Get Your API Key

Go to the API Keys page. A key is automatically generated for you. You'll need this to connect your AI assistant.

Step 6: Connect to Your AI Assistant

Add this to your MCP client configuration (e.g. Claude Desktop, Cursor, or any MCP-compatible client):

{
  "mcpServers": {
    "my-tools": {
      "url": "https://mcp.meetdolores.ai/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Replace YOUR_API_KEY with the key from the API Keys page. The exact config location depends on your client:

Claude Desktop — ~/.claude/claude_desktop_config.json
Cursor — Settings > MCP Servers

Google Sheets Setup

If your tool uses Google Sheets, you need a service account:

Go to Google Cloud Console
Create a project and enable the Google Sheets API
Go to APIs & Services > Credentials
Create a Service Account, then create a JSON key
Copy the client_email and private_key into your tool code
Share your Google Sheet with the service account email (give it Editor access)
The Sheet ID is in the URL: https://docs.google.com/spreadsheets/d/SHEET_ID/edit

Tips

Use AI Generate to write code from a plain English description
Use console.log() to debug — output shows in test results
Tool names must be letters, numbers, underscores, or hyphens only
The description helps the AI decide when to use your tool — be specific
Toggle the switch to enable/disable a tool without deleting it
Code runs in a sandbox with a 30-second timeout

Getting Started