{"type":"mcp_client","name":"tome","description":"**Option 1 (Focus on LLM client):**\n\nLLM desktop client leveraging MCP for easy access & use of Large Language Models.\n\n**Option 2 (Focus on MCP integration):**\n\nMCP-integrated desktop client simplify","category":"AI","language":"Svelte","stars":625,"forks":24,"owner":"runebookai","github_url":"https://github.com/runebookai/tome","homepage":"https://gettome.app","setup":"## Setup\n\n### Prerequisites\n\n*   **Operating System:** MacOS or Windows (Linux support is planned).\n*   **LLM Provider:** Choose one of the following:\n    *   [Ollama](https://ollama.com/): For running models locally.\n    *   [Gemini API key](https://aistudio.google.com/app/apikey): For using Google's Gemini models.\n\n### Installation\n\n1.  **Download Tome:** Get the latest release for your operating system from the [Tome Releases page](https://github.com/runebookai/tome/releases).\n    *   [Windows](https://github.com/runebookai/tome/releases/download/0.9.1/Tome_0.9.1_x64-setup.exe)\n    *   [MacOS](https://github.com/runebookai/tome/releases/download/0.9.1/Tome_0.9.1_aarch64.dmg)\n2.  **Install Tome:** Run the downloaded installer and follow the on-screen instructions.\n\n### Configuration\n\n1.  **Connect your LLM Provider:**\n    *   Open the Tome application.\n    *   Navigate to the settings or configuration panel (the specific location may vary within the app).\n    *   Configure your preferred LLM provider:\n        *   **Ollama:** If using Ollama, ensure it's running. Tome should automatically detect it.\n        *   **Gemini:** Enter your Gemini API key.\n        *   **Other OpenAI API-compatible endpoints (e.g., LM Studio):** Provide the API endpoint URL (e.g., `http://localhost:1234/v1`).\n2.  **Install an MCP Server:**\n    *   Open the MCP tab within Tome.\n    *   Enter the name of an MCP server in the server field (e.g., `uvx mcp-server-fetch`).\n    *   Click the install button.\n    *   Enable the MCP server.\n\n### Environment Variables\n\n*   No environment variables are explicitly required for basic setup. However, some MCP servers or LLM providers might require specific environment variables for authentication or configuration. Refer to the documentation for your chosen MCP server or LLM provider for details.","tools":"## Available Tools\n\n*   **Chat with Local or Remote LLMs:** Connect to various LLMs, either running locally (e.g., Ollama, LM Studio, Cortex) or remotely (e.g., Google Gemini, OpenAI, any OpenAI API-compatible endpoint).\n*   **Scheduled Tasks:** Schedule prompts to run automatically on an hourly or daily basis. This allows for automated workflows and background processing using LLMs.\n    *   *Example:* Schedule a daily summary of news articles or a recurring task to update a database.\n*   **Model Context Protocol (MCP) Support:** Integrates with MCP servers, providing access to external tools and data sources for LLMs.\n    *   UI to install, remove, turn on/off MCP servers\n    *   npm, uvx, node, python MCP servers all supported out of box\n*   **Smithery.ai Registry Integration:** Access thousands of MCP servers through the Smithery.ai registry for one-click installation.\n*   **Context Window and Temperature Customization:** Adjust the context window size and temperature settings to fine-tune the behavior of the LLM.\n*   **Native Tool Calls and Reasoning Model Support:** Enhanced UI to clearly display tool calls and reasoning processes performed by the LLM.","faq":null,"created_at":"2025-04-25T19:43:08+00:00","updated_at":"2025-08-07T17:00:08+00:00","source_url":"https://model-context-protocol.com/clients/tome","related_articles":[]}