Namecheap MCP – Domain Search Server

A Model Context Protocol (MCP) server that provides domain search, fuzzy domain suggestions, and availability lookup using Fastly Domain Research API, with Namecheap buy links for quick purchase.

This is designed to be consumed by MCP-compatible clients (e.g., Postman MCP, AI agents).

🚀 Features

• Fuzzy domain search from natural language queries
  → "apple", "cool ai studio", "apple.com" etc.
• Exact match detection
  → apple.com, apple.io, apple.ai
• Similar / brandable variations
  → getapple.com, tryapple.io, applehq.ai, etc.
• Live domain availability via Fastly Domain Research API
• Marks premium domains when detected
• Provides direct Namecheap buy link for every suggestion
• Fully async + fast response time

🧠 How it works (high-level)

Fastly API is used only to determine availability + premium status.
Prices are not included yet, but can be added later (Porkbun, Namecheap API, etc.).

📂 Project Structure

namecheap-mcp/
 ├─ src/
 │   └─ namecheap_mcp/
 │       ├─ server.py
 │       ├─ config.py
 │       ├─ controllers/
 │       ├─ services/
 │       ├─ repositories/
 │       ├─ models/
 ├─ pyproject.toml
 ├─ .env.example
 ├─ README.md

🔧 Requirements

• Python 3.12+
• uv (recommended) or Poetry
• MCP-compatible client (Postman MCP recommended)

📥 Setup

1️⃣ Install dependencies

uv sync

2️⃣ Create .env file in the project root:

FASTLY_API_TOKEN=your_fastly_api_token_here
MCP_HOST=0.0.0.0
MCP_PORT=8000

.env must NOT be committed. It is ignored via .gitignore.

3️⃣ Run the MCP server

uv run python -m namecheap_mcp.server

🧪 Testing With Postman MCP

1. Open Postman → Connect to MCP
2. Add connection:

http://localhost:8000

1. Call the tool:

search_domains

Example request:

{
  "query": "apple",
  "tlds": [".com", ".io", ".ai"],
  "max_results": 25,
  "include_taken": true
}

Response includes:

• exact_matches
• similar_matches
• available and is_premium
• fastly_status (raw API value)
• namecheap_buy_url