chatmcpFeatured

What is chatmcp?

Chatmcp is a cross-platform AI chat client that integrates with the Model Context Protocol (MCP) to allow users to chat with MCP servers. It runs on macOS, Windows, Linux, iOS, and Android, and supports multiple LLM models including OpenAI, Claude, Ollama, and DeepSeek.

How to use chatmcp?

After installing chatmcp, ensure you have uvx or npx installed on your system. Configure your LLM API key and endpoint in the Settings page, then install an MCP server from the MCP Server page (supports stdio and SSE transports). Finally, start chatting with the MCP server. Installation details (brew, curl, apt) are provided per platform.

Key features of chatmcp

• Chat with MCP servers via stdio or SSE transport
• Supports OpenAI, Claude, Ollama, and DeepSeek LLM models
• Built-in chat history and automatic MCP server selection
• Dark / Light theme
• Artifact display, thinking mode, and image generation
• LaTeX, HTML preview, and Mermaid diagram rendering

Use cases of chatmcp

• Connect to custom MCP servers to query local databases or APIs via natural language
• Rapidly prototype and test MCP server interactions with different LLM backends
• Use chatmcp as a unified AI assistant that can run on multiple devices (desktop and mobile)
• Leverage MCP server market to discover and install pre-built data connectors

FAQ from chatmcp

What platforms does chatmcp support?

Chatmcp supports macOS, Windows, Linux, iOS, and Android.

Which MCP transport protocols are supported?

Both stdio and SSE (Server-Sent Events) transport protocols are supported.

Does chatmcp require any system dependencies?

Yes, Linux requires libsqlite3-0 and libsqlite3-dev. On all platforms, uvx or npx must be available.

What is the license for chatmcp?

Chatmcp is licensed under the Apache License 2.0.

Are all features implemented?

Some features like MCP Server Market, auto install of MCP servers, RAG, and better UI are still in development (marked as unchecked). See the README for the full list.