Tome is a desktop application designed to simplify the use of Large Language Models (LLMs) and the Model Context Protocol (MCP) for a broad audience, from beginners to experienced users. It provides a user-friendly interface for interacting with local or remote LLMs and connecting them to various tools (MCP servers) for automated tasks and enhanced capabilities, aiming to make LLM technology more accessible and integrated into daily workflows.

How It Works

Tome acts as a central hub, abstracting away the complexities of LLM and MCP server management. Users can connect to diverse LLM providers, including local options like Ollama and remote services like OpenAI and Gemini. The application features a dedicated UI for discovering, installing, and managing MCP servers from the Smithery.ai registry, enabling LLMs to access external tools such as search engines and APIs. This approach prioritizes ease of use, allowing users to leverage LLM capabilities without deep technical configuration.

Quick Start & Requirements

• Install: Download the latest release for Windows or macOS.
• Prerequisites: An LLM provider (e.g., Ollama installed locally, or an API key for Gemini/OpenAI).
• Setup: Install Tome, connect an LLM provider, and install an MCP server (e.g., uvx mcp-server-fetch).
• Demo: https://github.com/user-attachments/assets/0775d100-3eba-4219-9e2f-360a01f28cce

Highlighted Details

• Supports local (Ollama, LM Studio) and remote (OpenAI, Gemini) LLM providers.
• Features a UI for installing and managing MCP servers from the Smithery.ai registry.
• Includes scheduled task functionality for automated LLM prompts.
• Offers native support for tool calls and reasoning models with clear UI delineation.

Maintenance & Community

• Community: Discord, Blog, Bluesky, Twitter.
• The project is described as a "Technical Preview," indicating ongoing development and potential instability.

Licensing & Compatibility

• The README does not explicitly state a license.

Limitations & Caveats

• Currently in "Technical Preview," meaning it may have rough edges and bugs.
• Linux support is listed as "coming soon."
• The absence of a stated license may pose compatibility concerns for commercial or closed-source use.