mcp-server-domain-tools
# mcp-server-domain-tools
[](https://www.npmjs.com/package/mcp-server-domain-tools)
[](https://opensource.org/licenses/MIT)
[](https://modelcontextprotocol.io)
Model Context Protocol (MCP) server providing an all-in-one Swiss Army knife suite for domain investors, webmasters, and autonomous agents.
---
## ⚡ Features & Tools
- **`full_domain_audit`**: Performs comprehensive domain audit including brand score, TLD tier analysis, and DNS connectivity checks.
- **`generate_brand_variations`**: Generates strategic brand suffixes and industry compound variations for portfolio expansion.
---
## 🚀 Quick Start (Running with AI Clients)
### Claude Desktop Integration
Add this snippet to your `claude_desktop_config.json`:
```json
{
"mcpServers": {
"domain-tools": {
"command": "npx",
"args": ["-y", "mcp-server-domain-tools"]
}
}
}
```
### Cursor IDE / Windsurf Integration
Under MCP Settings, add a new stdio transport:
- **Command:** `npx -y mcp-server-domain-tools`
---
## 📦 Local Installation & Development
```bash
npm install -g mcp-server-domain-tools
```
Or run directly via stdio:
```bash
npx mcp-server-domain-tools
```
---
## 📄 License
MIT License. Developed for the global AI agent and domain ecosystem.
TDQS
Scored across 2 tools
The two tools have fairly distinct purposes: one performs an audit/analysis of a domain, the other generates new brand name variations. There is slight conceptual overlap since both operate on domain/brand data, but an agent can tell them apart.
Both names use consistent snake_case and are readable. However, 'full_domain_audit' is a noun phrase while 'generate_brand_variations' follows a verb_noun pattern, so the convention isn't perfectly uniform.
Only 2 tools is thin for a domain-tools server. The pairing (audit + generate) is coherent, but the surface feels minimal for a domain-focused toolkit.
Key domain operations are absent: no availability/whois check, no TLD listing, no DNS-only lookup, and no way to inspect or manage results. The audit covers scoring and connectivity, but agents would hit dead ends for common domain tasks.