Databar MCP Server

A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with Databar.ai's data enrichment API. Discover, configure, and run data enrichments across hundreds of data providers using natural language.

Features

• Smart Enrichment Discovery — Search and filter enrichments by keyword or category

• Natural Language Interface — Ask "get David's LinkedIn profile" and the right enrichment runs automatically

• Bulk Operations — Enrich many records in a single call with bulk enrichment and bulk waterfall support

• Table Management — Create tables, manage columns, insert/update/upsert rows

• Waterfall Support — Try multiple data providers sequentially until one succeeds

• Async Handling — Automatic polling for results with no manual intervention

• Intelligent Caching — 24-hour result cache reduces API calls and costs

• Error Handling — Retries with exponential backoff and clear error messages

Quick Start

Prerequisites

• Node.js 18+

• A Databar.ai API key (get one here)

Install & Build

git clone https://github.com/databar-ai/databar-mcp-server.git
cd databar-mcp-server
npm install
npm run build

Configure Claude Desktop

Edit your Claude Desktop config file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "databar": {
      "command": "node",
      "args": ["/absolute/path/to/databar-mcp-server/dist/index.js"],
      "env": {
        "DATABAR_API_KEY": "your-api-key-here"
      }
    }
  }
}

Restart Claude Desktop. Verify by asking: "What Databar tools do you have access to?"

Usage Examples

Find someone's LinkedIn profile

"Get me David Abaev's LinkedIn profile"

Claude searches for LinkedIn enrichments, picks the right one, runs it, and returns the profile data.

Verify an email address

"Verify the email david@databar.ai"

Find an email using waterfall

"Find the email for John Smith at Google"

Runs a waterfall that tries multiple providers until one returns a result.

Bulk enrich a list

"Enrich these 10 emails with company data: [list]"

Uses bulk enrichment to process all records in a single API call.

Manage table data

"List my tables"
"Create 5 rows in table abc-123 with columns name and email"
"Get the columns for table abc-123"

Available Tools

Enrichments

Waterfalls

Tables

Row Operations

Account

Configuration

All settings are configurable via environment variables:

How It Works

Async Task Handling

1. Server sends a run request to the Databar API

2. API returns a task_id

3. Server automatically polls /v1/tasks/{task_id} every 2 seconds

4. When status is completed, results are returned

5. If data has expired (1-hour retention), gone status is handled gracefully

Caching

• Results are cached for 24 hours by default

• Cache key: enrichment ID + serialized params

• Cached results don't consume credits

• Use skip_cache: true to force fresh data

Smart Categorization

Enrichments are automatically categorized (People, Company, Email, Phone, Social, Financial, Verification) to help the AI assistant pick the right tool.

Development

npm run dev      # Run with tsx (hot reload)
npm run build    # Compile TypeScript
npm start        # Run compiled output

Project Structure

databar-mcp-server/
├── src/
│   ├── index.ts           # MCP server entry point & tool handlers
│   ├── databar-client.ts  # Databar API client with polling
│   ├── cache.ts           # In-memory cache with TTL
│   ├── types.ts           # TypeScript type definitions
│   └── utils.ts           # Helpers & categorization
├── dist/                  # Compiled output (generated)
├── package.json
├── tsconfig.json
└── .gitignore

Troubleshooting

Resources

• Databar API Documentation

• Databar Web App

• MCP Protocol Docs

License

MIT