What is the UBOS MCP Server?

The UBOS MCP Server is a Model Context Protocol server that provides real-time web and image search capabilities through Google’s Custom Search API, enhancing AI assistants’ access to current information.

How does the MCP Server integrate with AI assistants?

The MCP Server integrates with AI assistants by providing tools like google_web_search and google_image_search, which can be called by the AI to perform searches and access information.

What are the prerequisites for setting up the MCP Server?

To set up the MCP Server, you need Node.js (v18 or higher), a Google Cloud Platform account, a Google Custom Search API key, and a Search Engine ID.

How can I connect the MCP Server to MCP clients?

The server can be connected to MCP clients like Claude for Desktop, Cursor, and VSCode by configuring the client’s settings with the server’s command and environment variables.

What are the limitations of the Google Custom Search API?

The free tier of the Google Custom Search API allows 100 queries per day, with a server-enforced rate limit of 5 requests per second and a maximum of 10 results per query.