MCP 服务器

MarketIntel MCP Server

Enables AI assistants to perform real-time market research including competitor analysis, pricing intelligence, and company overviews via live web search through Tavily API.

README

🚀 MarketIntel MCP Server

An AI-powered Market Research MCP (Model Context Protocol) Server built using FastMCP, Python, Tavily Search API, and Cursor AI. This project enables Large Language Models (LLMs) to access real-time market intelligence through reusable MCP tools, providing structured competitor analysis, pricing insights, product portfolio mapping, and company research.

📌 Project Overview

MarketIntel is a custom MCP server that exposes market research capabilities as reusable tools. It integrates with the Tavily Search API to retrieve live web data and allows AI assistants (such as Cursor AI) to generate structured market intelligence reports.

The project demonstrates how Model Context Protocol (MCP) enables AI applications to securely interact with external services while maintaining a standardized interface.

✨ Features

📊 Company Overview
🏢 Competitor Analysis
📦 Product Portfolio Mapping
💰 Pricing Intelligence
📰 Recent News Monitoring
📈 SWOT & Porter's Five Forces Prompt
🌐 Live Web Search using Tavily
🤖 Cursor AI MCP Integration
⚡ FastMCP Server using SSE Transport

Architecture

                    Cursor AI
                        │
                        │ MCP
                        ▼
              MarketIntel MCP Server
                        │
         ┌──────────────┼──────────────┐
         │              │              │
         ▼              ▼              ▼
 Company Overview   Competitor     Pricing
                     Analysis      Intelligence
         │              │              │
         └──────────────┼──────────────┘
                        ▼
                  Tavily Search API
                        │
                        ▼
                 Live Web Search Results

Tech Stack

Python 3.12+
FastMCP
Tavily Search API
Cursor AI
uv Package Manager
Server-Sent Events (SSE)

Project Structure

MarketIntel-MCP/
│
├── server.py
├── .env
├── pyproject.toml
├── uv.lock
├── README.md
└── .gitignore

MCP Tools

Company Overview

Returns:

Company background
Headquarters
Products
Business model
Recent developments

Competitor Analysis

Returns:

Major competitors
Emerging competitors
Regional competitors
Market positioning

Product Portfolio

Maps:

Products
Solutions
Pricing tiers
Product categories

Pricing Snapshot

Retrieves:

Pricing
Billing models
Discounts
Regional pricing

Prerequisites

Install:

Python 3.12+
Cursor AI
uv
Tavily API Account

Installation

Clone the repository

git clone https://github.com/<yourusername>/MarketIntel-MCP.git

cd MarketIntel-MCP

Install dependencies

uv sync

uv add fastmcp
uv add tavily-python
uv add python-dotenv

Configure Environment Variables

Create a .env file.

TAVILY_API_KEY=your_api_key_here

Run the MCP Server

uv run server.py

Expected output

🚀 Starting MarketIntel MCP Server...

FastMCP Server running on

http://127.0.0.1:8000/sse

Configure Cursor AI

Open

Settings
→ Tools & Integrations
→ Add Custom MCP

Use

{
  "mcpServers": {
    "MarketIntel": {
      "url": "http://127.0.0.1:8000/sse"
    }
  }
}

Restart Cursor AI.

Example Prompt

Create a market research report comparing NVIDIA and AMD.

Cover:

• Company Overview
• Product Portfolio
• Pricing
• Competitors
• Recent News
• Future Outlook

Keep the report under 300 words.

Example Workflow

User Prompt
      │
      ▼
Cursor AI
      │
      ▼
MarketIntel MCP Server
      │
      ▼
FastMCP Tool
      │
      ▼
Tavily Search API
      │
      ▼
Live Market Data
      │
      ▼
Structured AI Report