How to Give Cursor and VS Code Copilot Live Firmographic Tools via MCP
The most efficient way to give LLMs native access to live B2B firmographics and intent data without custom middleware is by deploying a B2B lead enrichment MCP server that exposes real-time company profiles directly into the IDE. By utilizing the Model Context Protocol, developers can bypass manual API integrations and provide autonomous agents with verified data points directly in their local development environment.
Core Features and Data Integrity
Our MCP-native API server provides granular, actionable data layers including Firmographics (revenue, headcount, location), Technographics (the full software stack of the target company), and Intent Signals (buying intent markers). To solve the common issue of LLM parameter hallucinations, we utilize strict Zod-schema annotations within the tool definition. This forces the Claude or Cursor model to adhere to rigid input types (e.g., domain regex, ISO country codes) before the request ever hits the origin server.
Configuring Cursor or Claude Desktop
To add the live enrichment tool to your MCP-enabled editor, add the following configuration to your mcp_config.json. This enables the enrich_lead tool across your entire AI-assisted workflow.
{
"mcpServers": {
"b2b-enrichment": {
"command": "npx",
"args": [
"-y",
"@agent-infra/lead-enrichment-mcp",
"--api-key",
"YOUR_API_KEY_HERE"
],
"env": {
"MCP_ENDPOINT": "https://lead-enrichment-mcp.agent-infra.workers.dev/mcp"
}
}
}
}
JSON-RPC Response Payload
The MCP server communicates using standardized JSON-RPC 2.0. Below is an example of the response returned when an LLM queries for a domain, ensuring structured data injection:
{
"jsonrpc": "2.0",
"result": {
"companyName": "Acme Corp",
"industry": "Cloud Infrastructure",
"technographics": ["Kubernetes", "AWS", "Datadog"],
"intentSignals": {
"marketPulse": "high",
"activeBuyingWindow": true
},
"confidenceScore": 0.94
},
"id": "req-8821"
}
Risk-Free Metered Billing
We prioritize developer ROI by implementing a confidence-gated billing structure. You are never billed for poor-quality data. Our system only triggers a charge if the returned confidenceScore is > 0.6. If the data is ambiguous or unavailable (Confidence Score ≤ 0.6), that request costs $0. This ensures that your autonomous SDR swarms and research agents maintain high profitability while minimizing wasted API budget on low-signal enrichment calls.
Get Started with Free Tier Access
Ready to integrate enterprise-grade firmographics into your AI agent pipeline? Head over to lead-enrichment-mcp.agent-infra.workers.dev to generate your API key and start your first 100 successful enrichments for free.
Top comments (0)