MCP server for expired domains
Connect CatchDoms to Claude, Cursor, or any MCP-compatible assistant. Ask questions in plain English. Get domain data back.
Example conversation
What talking to your AI looks like
How it works
Connect once
Add CatchDoms to your MCP client config. Takes 30 seconds. Your API key handles authentication.
Ask questions
Talk to your AI normally. "Show me .de domains with DA above 30" or "What auctions end today?"
Get results
Your AI fetches the data, formats it, and can help you analyze or export it.
MCP Server for AI Assistants
Connect CatchDoms directly to Claude Code, Cursor, or any MCP-compatible AI assistant.
What is MCP?
Model Context Protocol (MCP) is an open standard that lets AI assistants securely connect to external data sources. Instead of making API calls manually, you can ask your AI assistant natural language questions like "find .fr domains older than 10 years with backlinks".
Setup
- Get your API key from the API dashboard
- Add this configuration to your MCP client
Claude Code: ~/.claude.json | Cursor: MCP settings
{
"mcpServers": {
"catchdoms": {
"type": "http",
"url": "https://catchdoms.com/mcp/catchdoms",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
Where to put this?
Claude Code: ~/.claude.json. Cursor: Settings > MCP. Other clients: check their docs for MCP server config.
Available tools
search_domains
Search domains with filters (source, TLD, score, age, backlinks, language, keywords)
get_domain
Get detailed information about a specific domain
list_filters
Get available filter options with counts
get_stats
Get platform statistics
Example prompts
MCP access = Pro plan
468 EUR/year. Includes REST API, CSV exports, expiry alerts, and MCP server access.
Questions
How is this different from the REST API?
Same data, different interface. The REST API is for code: you make HTTP requests, parse JSON. MCP is for conversation: you ask your AI, it handles the technical bits. Both use your API key, both count toward rate limits.
Which MCP client should I use?
Claude Code if you work in terminal. Cursor if you prefer a full IDE. Both support MCP servers out of the box. Any client that implements the MCP spec will work.
What are the rate limits?
60 requests per minute. Each AI tool call counts as one request. If your AI makes 5 searches in a conversation, that's 5 requests.
Is my API key safe?
Your key stays in your local config file. It's sent via HTTPS headers. We never see it in logs. Revoke it anytime from the API dashboard.