Model Context Protocol (MCP) Integration

Connect SearchHive's web scraping and search capabilities to any AI agent

Universal AI Agent Support

MCP is an open standard that allows AI agents to seamlessly integrate with external tools and data sources. Think of it as USB for AI integrations - one protocol, endless possibilities.

Quick Start

1. Install MCP Server

Install globally

cURL

npm install -g @searchhive/mcp-server

2. Configure API Key

Set environment variable

cURL

export SEARCHHIVE_API_KEY=sk_live_your_key_here

Get your API key from the dashboard

3. Connect to AI Client

Add SearchHive to your AI client's MCP configuration

Available Tools

SwiftSearch

swift_search

Real-time web search with optional auto-scraping of top results

Features

Real-time search

Auto-scraping

Contact extraction

Social discovery

Example Usage

swift_search({
  query: "latest AI developments 2025",
  auto_scrape_top: 3,
  include_contacts: true
})

ScrapeForge

scrape_forge

Enterprise web scraping with browser rendering

Features

JavaScript rendering

Bot detection bypass

Custom extraction

Timeout control

Example Usage

scrape_forge({
  url: "https://example.com",
  extract: ["title", "text", "links"],
  use_browser: true
})

DeepDive

deep_dive

AI-powered research across multiple sources

Features

Multi-source research

AI summarization

Social mentions

Citation tracking

Example Usage

deep_dive({
  query: "climate change impact 2025",
  max_sources: 10,
  generate_summary: true
})

Extract Contacts

extract_contacts

Extract emails and phone numbers from text

Features

Email detection

Phone extraction

Pattern matching

Source tracking

Example Usage

extract_contacts({
  text: "Contact us at hello@example.com or call (555) 123-4567"
})

Batch Scrape

batch_scrape

Scrape multiple URLs efficiently in batch

Features

Bulk processing

Error handling

Contact aggregation

Progress tracking

Example Usage

batch_scrape({
  urls: ["https://site1.com", "https://site2.com"],
  include_contacts: true
})

AI Client Setup

🤖

Claude Desktop

Supported

Native integration with Anthropic's Claude Desktop app

Setup: Add to MCP settings

⚡

Cursor

Supported

AI-powered code editor with MCP support

Setup: Configure in settings

🔄

Continue.dev

Supported

Open-source AI coding assistant

Setup: Add to config.json

📝

VS Code (MCP Extension)

Coming Soon

Visual Studio Code with MCP extension

Setup: Install extension

Claude Desktop Configuration

Add this configuration to your Claude Desktop MCP settings file:

claude_desktop_config.json

JSON

{
  "mcpServers": {
    "searchhive": {
      "command": "npx",
      "args": ["@searchhive/mcp-server"],
      "env": {
        "SEARCHHIVE_API_KEY": "sk_live_your_key_here"
      }
    }
  }
}

Cursor Configuration

Add this to your Cursor settings:

cursor-settings.json

JSON

{
  "mcp": {
    "servers": {
      "searchhive": {
        "command": "searchhive-mcp",
        "env": {
          "SEARCHHIVE_API_KEY": "sk_live_your_key_here"
        }
      }
    }
  }
}

Usage Examples

Research Assistant

Ask your AI agent to research any topic comprehensively:

💬 You:

"Research the latest developments in quantum computing and provide me with a comprehensive summary including key players and recent breakthroughs."

🤖 AI Agent:

Uses deep_dive tool to search multiple sources, scrape content, and generate an AI-powered summary with citations.

Lead Generation

Find and extract contact information from company websites:

💬 You:

"Find contact information for the top 5 AI startups in San Francisco."

🤖 AI Agent:

Uses swift_search to find companies, then batch_scrape with contact extraction to gather emails and phone numbers.

Content Analysis

Analyze and extract specific information from web pages:

💬 You:

"Scrape this product page and extract all the technical specifications and pricing information."

🤖 AI Agent:

Uses scrape_forge with custom extraction patterns to pull structured data from complex pages.

Troubleshooting

Authentication Issues

If you're getting authentication errors:

Verify your API key is correct and active in the dashboard
Ensure the SEARCHHIVE_API_KEY environment variable is set
Check that your plan has sufficient credits

Connection Issues

If the MCP server won't connect:

Restart your AI client after configuration changes
Check that the MCP server is properly installed: npx @searchhive/mcp-server --version
Verify the configuration syntax in your AI client settings

Tool Errors

If tools are failing:

Check the SearchHive API status for any outages
Verify URLs are accessible and properly formatted
Reduce batch sizes if hitting rate limits

Need Help?

Our team is here to help you get the most out of SearchHive's MCP integration