Manage AI Client Servers Effortlessly

MCP Linker is a cross-platform desktop app for adding and syncing MCP server configurations across AI clients without any LLM calls of its own - it's a configuration manager, not an MCP server that AI models call. It ships a built-in marketplace of 600+ curated MCP servers (including examples like sequential-thinking and desktop-commander); browsing the marketplace, clicking "Get" to view a server's configuration, and clicking "Add" installs it directly into a selected client. Local Sync keeps MCP server configs consistent across multiple clients on the same machine, Smart Detection auto-detects existing Python, Node.js, and uv environments so servers can be wired up without manual path configuration, and an optional Pro tier adds encrypted Cloud Sync across devices. It also bundles a GUI for the OpenAI Codex CLI, based on the related codexia project.

Use this when managing MCP server configurations across several AI clients at once - Claude Desktop, Claude Code, Cursor, VS Code, Cline, Windsurf, Codex, or Roo Code - and wanting a marketplace-driven, one-click way to discover and install servers rather than hand-editing each client's config file.

Multi-client support beyond a single client requires the paid plan, and Cloud Sync across devices is a Pro-tier feature; the base app covers local, single-machine sync. The project is dual-licensed: AGPL-3.0 for open-source use, but a separate Commercial License is required for closed-source, proprietary, hosted, or SaaS use - a distinction worth checking before embedding it in a commercial product.

A built-in marketplace of 600+ curated MCP servers with one-click add/configure/install into a chosen client. Local Sync propagates server configs across multiple installed clients. Smart Detection identifies existing Python, Node.js, and uv environments automatically. Cloud Sync (Pro) provides optional encrypted sync of configurations across devices. A dedicated GUI wraps the OpenAI Codex CLI for users who want a graphical front end to it.

On Arch Linux, install via the AUR with yay -S mcp-linker-bin. On macOS, use Homebrew: brew tap milisp/mcp-linker then brew install --cask mcp-linker. On Windows, Linux, or macOS generally, download the latest release directly from the project's GitHub Releases page. After installation, browse the built-in marketplace, click "Get" on a server to see its configuration, and click "Add" to install it into the selected AI client - if you have an active subscription, relaunch the app after first login for it to take effect. The project is dual-licensed under AGPL-3.0 and a separate Commercial License.

Users running multiple AI clients (Claude, Cursor, VS Code, Cline, Windsurf, Codex, Roo Code) who want a single place to discover, install, and keep MCP server configurations in sync, rather than manually editing each client's config file.

brew tap milisp/mcp-linker
brew install --cask mcp-linker