Manage AI Client Servers Effortlessly

MCP Linker is a cross-platform desktop app (macOS, Windows, Linux) for one-click adding and syncing Model Context Protocol server configurations across multiple AI clients, backed by a built-in marketplace and the official MCP registry (registry.modelcontextprotocol.io) - no LLM calls required to do the matching or installation.

Use it when managing MCP server configuration across several AI clients by hand - editing each client's own config file - has become tedious, or when discovering and installing new MCP servers from a curated/official registry is the goal. It supports Claude Desktop, Claude Code, Cursor, VS Code, Cline, Windsurf, Codex, and Roo Code as target clients, and can auto-detect Python, Node.js, and uv environments needed by a given server. It is not itself an MCP server to add to a client's config - it is the tool that writes those configs for you. For closed-source, proprietary, hosted, or SaaS use, the AGPL-3.0 license does not apply and a commercial license is required instead.

• Local Sync: keep MCP server configs consistent across multiple installed AI clients.
• Official MCP Registry browsing and one-click install straight from registry.modelcontextprotocol.io.
• Multi-client support across the eight clients listed above.
• Smart Detection of Python, Node.js, and uv runtime environments a server needs.
• A GUI for the OpenAI Codex CLI, built on the same team's Codexia project (file tree, notepad, git diff, plan mode in a lightweight Tauri app).

yay -S mcp-linker-bin

That AUR package covers Arch Linux; macOS installs via Homebrew (brew install --cask milisp/mcp-linker/mcp-linker), and Windows, Linux, and macOS all support a direct download from the project's GitHub Releases page. After installing, browse the built-in marketplace, click "Get" on a server to see its configuration, then click "Add" to install it into a selected client.

Developers running more than one AI coding client (Claude Desktop/Code, Cursor, VS Code, Cline, Windsurf, Codex, Roo Code) who want a single place to discover, install, and keep MCP server configuration in sync, without hand-editing each client's config file separately. The project describes Model Context Protocol itself as "like USB-C for AI" - a standard way to connect AI models to different data sources and tools - and MCP Linker is built specifically to make adopting that standard a one-click action rather than a manual JSON-editing exercise. The same maintainer also publishes related projects worth knowing about: Codexia, a GUI for the OpenAI Codex CLI that MCP Linker's own Codex GUI is based on; awesome-claude-dxt, a curated list of Claude Desktop Extensions; and a Python subprocess bridge for the Claude Code CLI. Support runs through GitHub Discussions and Issues, with a wiki for troubleshooting.