1. Clone/Place Project: Ensure the mcp-gemini-server project directory is accessible on your system.
2. Install Dependencies: Navigate to the project directory in your terminal and run:

    npm install
    

3. Build Project: Compile the TypeScript source code:

    npm run build
    

This command uses the TypeScript compiler (tsc) and outputs the JavaScript files to the ./dist directory (as specified by outDir in tsconfig.json). The main server entry point will be dist/server.js.
4. Generate Connection Token: Create a strong, unique connection token for secure communication between your MCP client and the server. This is a shared secret that you generate and configure on both the server and client sides.

Generate a secure token using one of these methods:

Option A: Using Node.js crypto (Recommended)

    node -e "console.log(require('crypto').randomBytes(32).toString('hex'))"

    

Option B: Using OpenSSL

    openssl rand -hex 32

    

Option C: Using PowerShell (Windows)

    [System.Convert]::ToBase64String([System.Security.Cryptography.RandomNumberGenerator]::GetBytes(32))

    

Option D: Online Generator (Use with caution)
Use a reputable password generator like 1Password or Bitwarden to generate a 64-character random string.

Important Security Notes:
- The token should be at least 32 characters long and contain random characters
- Never share this token or commit it to version control
- Use a different token for each server instance
- Store the token securely (environment variables, secrets manager, etc.)
- Save this token - you'll need to use the exact same value in both server and client configurations

5. Configure MCP Client: Add the server configuration to your MCP client's settings file (e.g., cline_mcp_settings.json for Cline/VSCode, or claude_desktop_config.json for Claude Desktop App). Replace /path/to/mcp-gemini-server with the actual absolute path on your system, YOUR_API_KEY with your Google AI Studio key, and YOUR_GENERATED_CONNECTION_TOKEN with the token you generated in step 4.

    {
      "mcpServers": {
        "gemini-server": { // Or your preferred name
          "command": "node",
          "args": ["/path/to/mcp-gemini-server/dist/server.js"], // Absolute path to the compiled server entry point
          "env": {
            "GOOGLE_GEMINI_API_KEY": "YOUR_API_KEY",
            "MCP_SERVER_HOST": "localhost",       // Required: Server host
            "MCP_SERVER_PORT": "8080",            // Required: Server port  
            "MCP_CONNECTION_TOKEN": "YOUR_GENERATED_CONNECTION_TOKEN", // Required: Use the token from step 4
            "GOOGLE_GEMINI_MODEL": "gemini-1.5-flash", // Optional: Set a default model
            // Optional security configurations removed - file operations no longer supported
            "ALLOWED_OUTPUT_PATHS": "/var/opt/mcp-gemini-server/outputs,/tmp/mcp-gemini-outputs" // Optional: Comma-separated list of allowed output directories for mcpCallServerTool and writeToFileTool
          },
          "disabled": false,
          "autoApprove": []
        }
        // ... other servers
      }
    }
    

Important Notes:
- The path in args must be the absolute path to the compiled dist/server.js file
- MCP_SERVER_HOST, MCP_SERVER_PORT, and MCP_CONNECTION_TOKEN are required unless NODE_ENV is set to test
- MCP_CONNECTION_TOKEN must be the exact same value you generated in step 4
- Ensure the path exists and the server has been built using npm run build
6. Restart MCP Client: Restart your MCP client application (e.g., VS Code with Cline extension, Claude Desktop App) to load the new server configuration. The MCP client will manage starting and stopping the server process.

Here are examples of how an MCP client (like Claude) might call these tools using the use_mcp_tool format:

Example 1: Simple Content Generation (Using Default Model)

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "prompt": "Write a short poem about a rubber duck."
    }
  </arguments>
</use_mcp_tool>

Example 2: Content Generation (Specifying Model & Config)

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-1.0-pro",
      "prompt": "Explain the concept of recursion in programming.",
      "generationConfig": {
        "temperature": 0.7,
        "maxOutputTokens": 500
      }
    }
  </arguments>
</use_mcp_tool>

Example 2b: Content Generation with Thinking Budget Control

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-1.5-pro",
      "prompt": "Solve this complex math problem: Find all values of x where 2sin(x) = x^2-x+1 in the range [0, 2π].",
      "generationConfig": {
        "temperature": 0.2,
        "maxOutputTokens": 1000,
        "thinkingConfig": {
          "thinkingBudget": 8192
        }
      }
    }
  </arguments>
</use_mcp_tool>

Example 2c: Content Generation with Simplified Reasoning Effort

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-1.5-pro",
      "prompt": "Solve this complex math problem: Find all values of x where 2sin(x) = x^2-x+1 in the range [0, 2π].",
      "generationConfig": {
        "temperature": 0.2,
        "maxOutputTokens": 1000,
        "thinkingConfig": {
          "reasoningEffort": "high"
        }
      }
    }
  </arguments>
</use_mcp_tool>

Example 3: Starting and Continuing a Chat

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_startChat</tool_name>
  <arguments>
    {}
  </arguments>
</use_mcp_tool>

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_sendMessage</tool_name>
  <arguments>
    {
      "sessionId": "some-uuid-123",
      "message": "Hello! Can you tell me about the Gemini API?"
    }
  </arguments>
</use_mcp_tool>

Example 4: Content Generation with System Instructions (Simplified Format)

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-2.5-pro-exp",
      "prompt": "What should I do with my day off?",
      "systemInstruction": "You are a helpful assistant that provides friendly and detailed advice. You should focus on outdoor activities and wellness."
    }
  </arguments>
</use_mcp_tool>

Example 5: Content Generation with System Instructions (Object Format)

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-1.5-pro-latest",
      "prompt": "What should I do with my day off?",
      "systemInstruction": {
        "parts": [
          {
            "text": "You are a helpful assistant that provides friendly and detailed advice. You should focus on outdoor activities and wellness."
          }
        ]
      }
    }
  </arguments>
</use_mcp_tool>

Example 6: Using Cached Content with System Instruction

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "modelName": "gemini-2.5-pro-exp",
      "prompt": "Explain how these concepts relate to my product?",
      "cachedContentName": "cachedContents/abc123xyz",
      "systemInstruction": "You are a product expert who explains technical concepts in simple terms."
    }
  </arguments>
</use_mcp_tool>

Example 6: Generating an Image

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateImage</tool_name>
  <arguments>
    {
      "prompt": "A futuristic cityscape with flying cars and neon lights",
      "modelName": "gemini-2.0-flash-exp-image-generation",
      "resolution": "1024x1024",
      "numberOfImages": 1,
      "negativePrompt": "dystopian, ruins, dark, gloomy"
    }
  </arguments>
</use_mcp_tool>

Example 6b: Generating a High-Quality Image with Imagen 3.1

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateImage</tool_name>
  <arguments>
    {
      "prompt": "A futuristic cityscape with flying cars and neon lights",
      "modelName": "imagen-3.1-generate-003",
      "resolution": "1024x1024",
      "numberOfImages": 4,
      "negativePrompt": "dystopian, ruins, dark, gloomy"
    }
  </arguments>
</use_mcp_tool>

Example 6c: Using Advanced Style Options

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateImage</tool_name>
  <arguments>
    {
      "prompt": "A futuristic cityscape with flying cars and neon lights",
      "modelName": "imagen-3.1-generate-003",
      "resolution": "1024x1024",
      "numberOfImages": 2,
      "stylePreset": "anime",
      "styleStrength": 0.8,
      "seed": 12345
    }
  </arguments>
</use_mcp_tool>

Example 7: Message Routing Between Models

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_routeMessage</tool_name>
  <arguments>
    {
      "message": "Can you create a detailed business plan for a sustainable fashion startup?",
      "models": ["gemini-1.5-pro", "gemini-1.5-flash", "gemini-2.5-pro"],
      "routingPrompt": "Analyze this message and determine which model would be best suited to handle it. Consider: gemini-1.5-flash for simpler tasks, gemini-1.5-pro for balanced capabilities, and gemini-2.5-pro for complex creative tasks.",
      "defaultModel": "gemini-1.5-pro",
      "generationConfig": {
        "temperature": 0.7,
        "maxOutputTokens": 1024
      }
    }
  </arguments>
</use_mcp_tool>

The response will be a JSON string containing both the text response and which model was chosen:

{
  "text": "# Business Plan for Sustainable Fashion Startup\n\n## Executive Summary\n...",
  "chosenModel": "gemini-2.5-pro"
}

Example 8: Using URL Context with Content Generation

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "prompt": "Summarize the main points from these articles and compare their approaches to sustainable technology",
      "urlContext": {
        "urls": [
          "https://example.com/sustainable-tech-2024",
          "https://techblog.com/green-innovation"
        ],
        "fetchOptions": {
          "maxContentKb": 150,
          "includeMetadata": true,
          "convertToMarkdown": true
        }
      },
      "modelPreferences": {
        "preferQuality": true,
        "taskType": "reasoning"
      }
    }
  </arguments>
</use_mcp_tool>

Example 9: Advanced URL Analysis

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_url_analysis</tool_name>
  <arguments>
    {
      "urls": ["https://company.com/about", "https://company.com/products"],
      "analysisType": "extraction",
      "extractionSchema": {
        "companyName": "string",
        "foundedYear": "number",
        "numberOfEmployees": "string",
        "mainProducts": "array",
        "headquarters": "string",
        "financialInfo": "object"
      },
      "outputFormat": "json",
      "query": "Extract comprehensive company information including business details and product offerings"
    }
  </arguments>
</use_mcp_tool>

Example 10: Multi-URL Content Comparison

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_url_analysis</tool_name>
  <arguments>
    {
      "urls": [
        "https://site1.com/pricing",
        "https://site2.com/pricing", 
        "https://site3.com/pricing"
      ],
      "analysisType": "comparison",
      "compareBy": ["pricing models", "features", "target audience", "value proposition"],
      "outputFormat": "markdown",
      "includeMetadata": true
    }
  </arguments>
</use_mcp_tool>

Example 11: URL Content with Security Restrictions

<use_mcp_tool>
  <server_name>gemini-server</server_name>
  <tool_name>gemini_generateContent</tool_name>
  <arguments>
    {
      "prompt": "Analyze the content from these trusted news sources",
      "urlContext": {
        "urls": [
          "https://reuters.com/article/tech-news",
          "https://bbc.com/news/technology"
        ],
        "fetchOptions": {
          "allowedDomains": ["reuters.com", "bbc.com"],
          "maxContentKb": 200,
          "timeoutMs": 15000,
          "userAgent": "Research-Bot/1.0"
        }
      }
    }
  </arguments>
</use_mcp_tool>

- GOOGLE_GEMINI_MODEL: Default model to use (e.g., gemini-1.5-pro-latest, gemini-1.5-flash)
- GOOGLE_GEMINI_DEFAULT_THINKING_BUDGET: Default thinking budget in tokens (0-24576) for controlling model reasoning