mcp-gemini-server

Overview

This MCP server provides a comprehensive interface to Google's Gemini models through the Model Context Protocol. It wraps the @google/genai SDK to expose Gemini capabilities as standard MCP tools, enabling other LLMs or MCP-compatible systems to leverage Gemini as a backend. The server supports multiple Gemini models including gemini-1.5-pro-latest, gemini-1.5-flash, and gemini-2.5-pro. It offers features like core text generation, function calling, stateful chat management, URL-based multimedia analysis, caching, and image generation. While it doesn't support direct file uploads, it excels at analyzing publicly accessible URLs including images and YouTube videos.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• gemini_generateContent

  Generates non-streaming text content from a prompt with optional URL context support.

  Try:Explain quantum computing in simple termsTry:Analyze the content at https://example.com/article
• gemini_generateContentStream

  Generates streaming text content from a prompt with optional URL context support.

  Try:Write a story about a robot discovering emotionsTry:Summarize the key points from https://example.com/research
• gemini_functionCall

  Enables Gemini models to request the execution of client-defined functions.

  Try:Help me plan a trip to Paris with a budget of $2000Try:Find the best restaurant in my area for Italian food
• gemini_startChat

  Starts a new chat session with Gemini for stateful conversations.

  Try:Start a new chat about climate changeTry:Begin a conversation about cooking techniques
• gemini_sendMessage

  Sends a message in an ongoing chat session with Gemini.

  Try:What's the next step in our climate discussion?Try:How do I make a perfect souffle?
• gemini_generateImage

  Generates images from text prompts using Gemini 2.0 Flash Experimental.

  Try:Create a painting of a futuristic city at sunsetTry:Generate a logo for a tech startup
• gemini_url_analysis

  Specialized tool for advanced URL content analysis with multiple analysis types.

  Try:Analyze this YouTube video for educational content: https://youtube.com/watch?v=exampleTry:Extract key information from this webpage: https://example.com/news
• mcpConnectToServer

  Establishes a connection to an external MCP server.

  Try:Connect to the MCP server at localhost:3000Try:Link to the GitHub issues MCP server
• mcpListServerTools

  Lists available tools on a connected MCP server.

  Try:Show me all tools available on the connected serverTry:What functions can I call on the GitHub server?
• mcpCallServerTool

  Calls a function on a connected MCP server, with an option for file output.

  Try:Create a new GitHub issue with title 'Bug report' and description 'App crashes on login'Try:Get the weather for New York City using the weather service
• mcpDisconnectFromServer

  Disconnects from an external MCP server.

  Try:Disconnect from the GitHub issues serverTry:End connection to the weather service
• writeToFile

  Writes content directly to files within allowed directories.

  Try:Save my essay to /documents/my_essay.txtTry:Export the analysis results to /reports/summary.md

Comparable tools

Installation

FAQ

Does this server support direct file uploads?

No, this MCP Gemini Server does not support direct file uploads. It focuses on URL-based multimedia analysis for images and videos. Local files must be hosted publicly to be analyzed.

Can I use this with Vertex AI credentials?

No, this server only supports Google AI Studio API keys. The Caching API is not compatible with Vertex AI credentials and is not currently supported.

Compare mcp-gemini-server with