Google Patents Mcp

What is Google Patents Mcp?

Google Patents Mcp is a Model Context Protocol (MCP) server that allows searching Google Patents information via the SerpApi Google Patents API. It is designed for developers and AI assistants who need to query patent data programmatically.

How to use Google Patents Mcp?

The server can be run directly using npx without local installation. Configuration requires a valid SerpApi API key, which can be supplied via the SERPAPI_API_KEY environment variable or a .env file. For MCP hosts, set the environment variable in the host configuration. After starting, the server listens for MCP requests on standard input/output.

Key features of Google Patents Mcp

• Provides the search_patents MCP tool for patent searches.
• Uses SerpApi as the backend for patent data.
• Can be executed directly via npx without installation.
• Supports filtering by date, inventor, assignee, country, language, status, and type.
• Allows inclusion of Google Scholar results.

Use cases of Google Patents Mcp

• Searching for granted patents in a specific language and date range.
• Filtering patents by inventor or assignee for competitive analysis.
• Integrating patent search capabilities into AI assistants via MCP.

FAQ from Google Patents Mcp

What are the prerequisites for using Google Patents Mcp?

Node.js version 18 or higher is recommended, along with npm and a valid SerpApi API key.

How do I configure the SerpApi API key?

You can set the SERPAPI_API_KEY environment variable (recommended for MCP hosts) or create a .env file (in the current directory or ~/.google-patents-mcp.env) with the key.

What tool does Google Patents Mcp provide?

It provides a single MCP tool called search_patents, which accepts a search query and optional filters (page, number of results, sort, date range, inventor, assignee, country, language, status, type, and scholar flag).

Can I run the server without installing it locally?

Yes, you can run it directly via npx @kunihiros/google-patents-mcp. The package is downloaded and executed on the fly.

What logging options are available?

Log level is controlled by the LOG_LEVEL environment variable (e.g., error, warn, info, debug). Logs are output to standard error and optionally written to a log file.