DetectZeStack MCP Server: Tech Detection for AI Agents (2026)

DetectZeStack is now available as an MCP server. If you use Claude Code, Cursor, or Windsurf, you can detect tech stacks, scan security headers, check SSL certificates, and run DNS lookups as native tools — directly from your AI coding assistant, without writing any code.

One command to install. Nine tools available immediately. Free tier included.

npx detectzestack-mcp

What Is MCP?

MCP (Model Context Protocol) is an open standard created by Anthropic that lets AI agents use external tools. Think of it as a plugin system for AI coding assistants. Instead of copying API responses and pasting them into your chat, MCP lets the AI call the tool directly, get structured results, and reason about the data.

Before MCP, if you wanted your AI assistant to check what tech stack a website uses, you had to:

1. Open a terminal
2. Run a curl command with API keys and headers
3. Copy the JSON response
4. Paste it into your AI conversation
5. Wait for the AI to parse and interpret the output

With MCP, you just say: “Detect the tech stack of stripe.com.” The AI calls the tool, gets the response, and gives you the answer. No copy-pasting, no terminal switching, no manual API calls.

What the DetectZeStack MCP Server Does

The server exposes 9 tools that map directly to the DetectZeStack API endpoints:

Every tool returns structured JSON. The AI agent parses it, summarizes the results, and can chain tools together — for example, detecting a tech stack and then scanning for vulnerabilities in the same conversation.

How to Install

You need two things: Node.js 18+ and a free API key from RapidAPI (no credit card required). The server also supports API.market and direct API keys.

Claude Code

Add to your project’s .mcp.json:

{
  "mcpServers": {
    "detectzestack": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "detectzestack-mcp"],
      "env": {
        "RAPIDAPI_KEY": "your-rapidapi-key"
      }
    }
  }
}

Cursor

Go to Settings → MCP Servers → Add Server, or add to .cursor/mcp.json:

{
  "mcpServers": {
    "detectzestack": {
      "command": "npx",
      "args": ["-y", "detectzestack-mcp"],
      "env": {
        "RAPIDAPI_KEY": "your-rapidapi-key"
      }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "detectzestack": {
      "command": "npx",
      "args": ["-y", "detectzestack-mcp"],
      "env": {
        "RAPIDAPI_KEY": "your-rapidapi-key"
      }
    }
  }
}

Example Prompts

Once the MCP server is connected, you can ask your AI assistant questions in natural language. Here are some examples:

Tech Stack Detection

"Detect the tech stack of stripe.com"
"What framework does vercel.com use?"
"Does shopify.com use React or Vue?"
"Compare the tech stacks of stripe.com, shopify.com, and square.com"

Security Analysis

"Check the security headers for my site at example.com"
"Grade the security headers of github.com"
"What SSL certificate does github.com use? Is it expiring soon?"
"Scan example.com for known vulnerabilities"

DNS & Infrastructure

"Look up DNS records for stripe.com"
"What email security does gmail.com have? Check SPF, DKIM, DMARC"
"Give me a full profile of example.com — tech, DNS, SSL, and security headers"

The AI handles parameter mapping, error handling, and result formatting automatically. You just ask the question.

Why This Matters

AI coding assistants are becoming the primary interface developers use to interact with tools. Instead of memorizing API endpoints, crafting HTTP requests, and parsing JSON, you describe what you want and the AI does the rest.

MCP makes this possible without custom integrations. The DetectZeStack MCP server works with any AI tool that supports the Model Context Protocol — today that includes Claude Code, Cursor, and Windsurf, with more adopting the standard regularly.

Some concrete workflows this enables:

• Security audits during code review — “Before I merge this PR, check the security headers on our staging site and scan for vulnerabilities.”
• Competitive research while building — “What framework does our competitor use? Compare their stack with ours.”
• Certificate monitoring — “Check if our SSL cert is expiring soon. When does it expire?”
• Infrastructure debugging — “Look up the DNS records for our domain. What are the MX records?”
• Vendor evaluation — “Find sites using Stripe. How many are in our database?”

All of this happens inside your editor, in the same conversation where you are writing code.

Lookup Result Limits by Plan

The lookup_by_technology tool (which finds domains using a specific technology) returns different numbers of results depending on your plan:

The site_profile tool is a premium endpoint that requires the Ultra or Mega plan and costs 3 API credits per call. All other tools cost 1 credit each (except compare_stacks, where each URL counts as 1 credit).

Source Code & npm Package

The MCP server is open source and published on npm:

• npm: npmjs.com/package/detectzestack-mcp
• GitHub: github.com/mlugo-apx/detectzestack-mcp

The server is MIT-licensed. The underlying DetectZeStack API that powers the tools is a commercial service with a free tier.