Lead Qualifier

You have a lightweight MCP tool that uses a conversational flow to qualify leads using the BANT model (Budget, Authority, Need, Timeline). It runs as an MCP server in Node.js, tracks lead information in memory for fast interactions, and can be connected to Dify or Cursor via an MCP SSE transport. You interact with it one question at a time, guiding users to provide the necessary data and scoring each dimension to produce a final qualification score.

How to use

To use the Lead Qualifier MCP Tool, start the server and connect an MCP client. The tool engages users in a turn‑by‑turn conversation, asking one field at a time (budget, authority, need, timeline) and updating the qualification score after each response. It outputs a structured session state and the next field to prompt, so you can build a smooth, guided qualification flow for your leads.

Key usage patterns you can implement with an MCP client include:

• Start a new session and present a welcome prompt that introduces the qualification process.
• Capture the lead’s budget in the first turn and score it against the budget dimension.
• Proceed to the authority, need, and timeline questions in sequence, updating scores and the overall total after each answer.
• Use the provided sample prompts to keep the flow natural and focused on discovering decision makers and timelines.

How to install

Prerequisites you need installed on your machine before running the server are Node.js and npm. Ensure you have a modern Node.js runtime (for example Node.js 18+). Then follow these steps to run the Lead Qualifier MCP Tool locally.

# Install dependencies for the MCP server
npm install

# Start the MCP server
npm start

Additional sections

Configuration and usage notes for the Lead Qualifier MCP Tool are shown below. You can set your OpenAI API key in a .env file to enable the LLM-backed qualification flow.

Environment variable for OpenAI API key (required for LLM features): you should place the key in a .env file at the project root.

OPENAI_API_KEY=YOUR_API_KEY

Optional: expose your MCP server via ngrok for quick external access during development.

ngrok http 3001

Dify Agent Strategy Configuration (example). This config enables the MCP server to be reachable via SSE transport through Dify or Cursor.

{
  "lead_qualification": {
    "transport": "sse",
    "url": "https://YOUR_NGROK_URL/sse",
    "headers": {},
    "timeout": 50,
    "sse_read_timeout": 50
  }
}

Available tools

lead_qualifier

Core MCP tool that conducts BANT-based lead qualification with a one-question-at-a-time conversational flow and in-memory session tracking.

sse_transport

Transport layer enabling SSE-based communication with compatible MCP clients like Dify or Cursor.