What is the Traylinx Search Engine MCP Server?

The Traylinx Search Engine MCP Server acts as a bridge between MCP clients (like Claude and Cursor) and the Traylinx Agentic Search API, providing intelligent search capabilities.

What is MCP (Model Context Protocol)?

MCP is an open protocol that standardizes how applications provide context to LLMs, enabling AI models to access and interact with external data sources and tools.

What kind of search recency filters are supported?

The server supports filtering search results by “month”, “week”, “day”, or “hour”.

What do I need to use the Traylinx Search Engine MCP Server?

You need a Traylinx.com API key, Node.js (>= 18.0.0), and an MCP-compatible client like Claude Desktop or Cursor.

Where do I configure the Agentic Search API Key?

You configure the API key in your MCP client’s configuration file (e.g., claude_desktop_config.json or mcp.json) as an environment variable named AGENTIC_SEARCH_API_KEY.

What content types does the server return?

The server can return text content, embedded HTML, structured search results, media items, news articles, and the raw API response.

How do I filter search results by recency?

You can specify the search_recency_filter argument in your search query with values like “month”, “week”, “day”, or “hour”.

What if I encounter issues during setup?

Double-check your API key, ensure the MCP client is fully restarted, verify network connectivity, and set LOG_LEVEL to DEBUG for more detailed logs.

Can I deploy this server to Smithery.ai?

Yes, you can deploy the server to Smithery.ai using the provided instructions.

Where can I get support?

For additional support, contact the API provider at support@traylinx.com.