DEV Community

Diego Costa
Diego Costa

Posted on

How to Connect Claude Desktop to a Live B2B Lead Enrichment MCP API

#mcp #ai #automation

How to Connect Claude Desktop to a Live B2B Lead Enrichment MCP API

Integrating real-time firmographics into your AI workflow is now possible via the Model Context Protocol (MCP), allowing Claude Desktop to pull live B2B lead enrichment data directly into your context window. By deploying a B2B lead enrichment MCP server, you can eliminate manual lookups and enable autonomous agents to access verified company profiles and technographic signals without leaving their native interfaces.

Core Features

This MCP server provides deep integration into high-fidelity data layers, ensuring your LLM agents operate on ground-truth data rather than stale training set information.

  • Firmographic Data: Access granular companyName, industry, employeeCount, and annualRevenue.
  • Technographic Stack: Identify software usage patterns, hosting providers, and CRM/marketing automation tools.
  • Intent Signals: Receive real-time scoring on market interest.
  • Zero-Hallucination Schema: All tool calls utilize strict Zod annotations within the MCP schema, forcing the LLM to provide standardized parameters for domain_name or company_name queries, effectively neutralizing common parameter hallucination issues.

Configuring Claude Desktop

To integrate the B2B Lead Enrichment MCP API into Claude Desktop, update your claude_desktop_config.json file. This setup enables the enrich_lead tool natively:

{
  "mcpServers": {
    "b2b-enrichment": {
      "command": "npx",
      "args": [
        "-y",
        "@agent-infra/lead-enrichment-mcp"
      ],
      "env": {
        "API_KEY": "YOUR_API_KEY_HERE",
        "MCP_ENDPOINT": "https://lead-enrichment-mcp.agent-infra.workers.dev/mcp"
      }
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

JSON-RPC Response Payload

When the LLM calls the enrich_lead tool, the server returns a structured JSON-RPC response designed for machine consumption. Note the confidenceScore, which acts as a gatekeeper for data quality:

{
  "jsonrpc": "2.0",
  "result": {
    "companyName": "Acme Corp",
    "industry": "SaaS",
    "technographics": ["AWS", "Salesforce", "Segment"],
    "intentSignals": {
      "buyingStage": "active_research",
      "score": 88
    },
    "confidenceScore": 0.94
  },
  "id": "1"
}
Enter fullscreen mode Exit fullscreen mode

The Value Proposition: Risk-Free Metered Billing

We prioritize cost-efficiency for developers building autonomous SDR swarms. Our pricing model is strictly metered by success: you are only billed for enrichments that return a Confidence Score > 0.6. If the API fails to find a match or returns low-quality data (Confidence Score <= 0.6), the query is free. This ensures your AI workflows are never charged for dead-ends or empty results.

Get Started Today

Ready to turn your LLM into a high-powered research assistant? Head over to lead-enrichment-mcp.agent-infra.workers.dev to grab your API key on the free tier and start enriching leads with native MCP tools in under five minutes.

Top comments (0)