MCP ready

MCP Server,
connect AI to website intelligence

Integrate Klazify with Claude Desktop, ChatGPT, or any MCP-compatible assistant. Classify URLs and enrich domains with categories, logos, social links, company data, and tech stacks.

Get API key Explore MCP tools

Ask Claude:

"What category is cbsnews.com, and what's their tech stack?"

AI calls MCP tool:

{

"tool": "klazify.categorize",

"params": {

"url": "https://cbsnews.com"

}

What is MCP and why does it matter?

The Model Context Protocol (MCP) lets AI assistants access external data in real time. Instead of outdated training data, your AI can fetch live website intelligence from Klazify — just by asking a question.

Example: Ask Claude "What category is cbsnews.com?" and it will call our klazify.categorize tool automatically.

Live enrichment

Categories, company, tech, logo & social from any URL

Works with Claude & ChatGPT

Native MCP plus an OpenAI-compatible proxy

Same API quota

Each tool call counts as one Klazify API request

Available MCP tools

Three tools mapped to categorize, company data, and logo endpoints. Pass api_key on every call.

klazify.categorize Core

Classify a URL or email domain and return categories (with confidence), logo, social media, company firmographics, and tech stack in one call. Maps to POST /api/categorize.

klazify.domain_company Company

Company intelligence: name, headquarters, employees, revenue, raised capital, industry tags, and tech stack. Maps to POST /api/domain_company.

klazify.domain_logo Brand

Hosted logo URL for a company/domain, ready to embed. Maps to POST /api/domain_logo.

Getting started

Connect in minutes using your API key in the endpoint URL. For direct programmatic calls, include api_key in the tool arguments.

MCP endpoint

https://mcp.klazify.com/mcp?api_key=YOUR_API_KEY

Replace YOUR_API_KEY with your key from the dashboard.

Claude Desktop setup

Add Klazify to Claude Desktop via mcp-remote.

Open Claude Desktop → Settings → Developer → Edit Config
Paste the JSON below into claude_desktop_config.json
Restart Claude Desktop

claude_desktop_config.json

{
  "mcpServers": {
    "klazify": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.klazify.com/mcp?api_key=YOUR_API_KEY"
      ]
    }
  }
}

After setup, try asking Claude: "What category is cbsnews.com?"

OpenAI proxy

Our OpenAI-compatible proxy enriches ChatGPT responses with live Klazify data. Send your request to our endpoint instead of OpenAI's.

How it works:

We fetch domain intelligence using the MCP tool you specify
That data is injected into the model context
OpenAI responds with grounded, up-to-date website data

curl — categorize cbsnews.com

curl -X POST "https://mcp.klazify.com/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_OPENAI_API_KEY" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"What category is cbsnews.com?"}],"force_mcp_tool":{"tool_name":"klazify.categorize","params":{"api_key":"YOUR_API_KEY","url":"https://cbsnews.com"}}}'

Authorization — your OpenAI API key (header only)
force_mcp_tool.tool_name — use klazify.categorize
force_mcp_tool.params.api_key — your Klazify API key
force_mcp_tool.params.url — e.g. https://cbsnews.com

Python client

A complete Python example that categorizes cbsnews.com using the MCP protocol.

python

import asyncio
import json
from mcp import ClientSession
from mcp.client.sse import sse_client

MCP_URL = "https://mcp.klazify.com/mcp?api_key=YOUR_API_KEY"

async def call_mcp_tool(tool_name: str, arguments: dict):
    async with sse_client(MCP_URL) as (read, write):
        async with ClientSession(read, write) as session:
            await session.initialize()
            result = await session.call_tool(tool_name, arguments)
            return result

async def main():
    result = await call_mcp_tool(
        "klazify.categorize",
        {
            "api_key": "YOUR_API_KEY",
            "url": "https://cbsnews.com",
        },
    )
    print(json.dumps(result.model_dump() if hasattr(result, "model_dump") else str(result), indent=2))

if __name__ == "__main__":
    asyncio.run(main())

Connect ChatGPT

Use Klazify as a custom MCP connector in ChatGPT (where supported).

1. Open ChatGPT settings

Go to Settings → Connectors / Advanced and add a custom MCP server.

2. Paste this MCP endpoint

https://mcp.klazify.com/mcp?api_key=YOUR_API_KEY

3. Try these prompts

"What category is cbsnews.com?"
"Get the company data for cbsnews.com"
"Get the logo for cbsnews.com"

FAQs

How do I get a Klazify API key?

An API key is required to use these tools. Start your free trial to get access. Check our pricing page for plan details. Pass the key as api_key in the tool arguments (the Worker sends it as Authorization: Bearer).

How much does each MCP request cost?

MCP requests count toward your existing API quota just like REST API calls. Each tool call equals one API request. Check our pricing page for plan details.

What does klazify.categorize return?

Categories with confidence scores, logo URL, social media links, company data (name, HQ, employees, tags), and detected technologies. See the categorize docs for the full response schema.

Is the MCP server compatible with other AI platforms?

Yes. Any platform that supports the Model Context Protocol (MCP) can connect — Claude Desktop, custom Python clients, ChatGPT connectors, and future MCP-compatible platforms.

MCP Server, connect AI to website intelligence