Serper MCP Server is a Model Context Protocol server that exposes a family of Google search tools powered by the Serper API. It enables LLMs to access and reason over search results by calling dedicated tools such as google_search, google_search_images, google_search_videos, google_search_places, google_search_maps, google_search_news, google_search_shopping, google_search_autocomplete, and more. Each tool accepts a rich set of parameters defined in the server's schemas, allowing fine-grained control over query, location, language, safe search, time range, result count, and other modifiers. The server is implemented in Python, requires Python 3.11+, and can be installed via pip or through Smithery for Claude Desktop usage. At runtime, the server expects a SERPER_API_KEY environment variable, and can be run with uv (recommended). It includes an MCP inspector for debugging. The project is MIT licensed. This server-level component acts as an intermediary between LLM clients and Google's search results, returning structured data suitable for inclusion in language-model reasoning and responses.