DEV Community

Diego Costa
Diego Costa

Posted on

How to Give Cursor and VS Code Copilot Live B2B Firmographic Tools via MCP

#mcp #ai #automation

How to Give Cursor and VS Code Copilot Live B2B Firmographic Tools via MCP

Giving your AI-powered IDEs like Cursor and VS Code native access to live B2B firmographics and intent data is best achieved by integrating a Model Context Protocol (MCP) server directly into your development environment. This setup allows LLMs to pull real-time company profiles and enrichment data using standardized JSON-RPC calls, eliminating the need for manual copy-pasting from web browsers or external CRM platforms.

Core Features

The B2B Lead Enrichment MCP server provides a high-fidelity data layer directly to your LLM's context window. By leveraging Zod-based schema definitions, the server forces the LLM to provide strictly typed arguments (e.g., valid domain formats or email syntaxes), which effectively eliminates parameter hallucinations common in unstructured sales tool calls.

  • Deep Firmographics: Access real-time data on company size, revenue brackets, headquarters location, and industry classification.
  • Technographic Intelligence: Identify the software stack and infrastructure used by a target company to tailor engineering-led growth strategies.
  • Intent Signaling: Retrieve live signals indicating high-purchase intent or recent funding rounds.
  • Strict Parameter Validation: Every tool (like enrich_lead and get_company_intent) utilizes strict JSON schema validation to ensure the LLM never "invents" search parameters.

Integration Configuration

To enable these tools in Cursor or Claude Desktop, add the following configuration to your settings. This connects your local agent to the production-grade MCP endpoint.

{
  "mcpServers": {
    "b2b-enrichment": {
      "command": "npx",
      "args": [
        "-y",
        "@agent-infra/mcp-server-b2b-enrichment",
        "--api-url",
        "https://lead-enrichment-mcp.agent-infra.workers.dev/mcp"
      ],
      "env": {
        "ENRICHMENT_API_KEY": "YOUR_FREE_API_KEY_HERE"
      }
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Example JSON-RPC Tool Response

When you ask Cursor, "Who is the CTO of stripe.com and what is their current stack?", the MCP server returns a structured response that the LLM uses to formulate its answer:

{
  "method": "enrich_lead",
  "result": {
    "companyName": "Stripe",
    "industry": "Fintech / Payments",
    "technographics": ["AWS", "React", "Ruby on Rails", "Kubernetes"],
    "intentSignals": [
      { "type": "Expansion", "score": 0.88, "description": "Increasing hiring for infrastructure roles" }
    ],
    "confidenceScore": 0.96,
    "source": "verified_firmographic_index"
  }
}
Enter fullscreen mode Exit fullscreen mode

Risk-Free Metered Billing

Unlike traditional B2B data providers that charge per query regardless of quality, this MCP server utilizes a Confidence-First Billing model. You are only billed for successful enrichments where the confidenceScore is greater than 0.6. If the data is unavailable, of low quality, or the query fails, the cost is exactly $0. This allows developers to build autonomous SDR swarms and agentic workflows that are highly cost-effective and scalable without worrying about burning credits on "Not Found" responses.

Get Your Free API Key Today

Ready to empower your AI agents with real-time B2B intelligence? Head over to the documentation and grab your API key to start building.

Visit: lead-enrichment-mcp.agent-infra.workers.dev

Top comments (0)