mcpx-cli
Interactive CLI to configure MCP servers for multiple AI providers
claude mcp add --transport stdio gustavodiasdev-mcpx-cli npx -y @gustavodiasdev/mcpx-cli
How to use
MCPX-CLI (mcpx-cli) is a command-line orchestrator that lets you manage a single canonical MCP configuration file (.mcpx.json) and generate per-provider config files for multiple AI CLIs. With mcpx-cli installed, you can run an interactive wizard to detect existing MCP configurations, import servers from those configs, add new MCP servers, select which providers to generate configs for, and then generate or synchronize all provider config files in one go. The tool centralizes your MCP definitions and automatically emits the corresponding files for each provider (e.g., .mcp.json for Claude Code, .gemini/settings.json for Gemini CLI, etc.). This makes it easier to maintain a single source of truth while supporting many AI CLIs across projects.
How to install
Prerequisites:
- Node.js >= 20
- npm (comes with Node.js) or Corepack enabled
Install MCPIX-CLI globally:
npm install -g mcpx-cli
Verify installation:
mcpx --version
Usage overview:
- From your project directory, run
mcpxto start the interactive wizard. - Follow prompts to detect existing configs, import servers, add new servers, and select providers.
- Run
mcpx syncto regenerate all provider configuration files based on your .mcpx.json.
If you prefer not to install globally, you can run via npx directly in the project:
npx -y mcpx-cli
Additional notes
Tips and notes:
- The canonical configuration file is .mcpx.json. Providers generate their own provider-specific config files from this.
- You can import existing configurations from various tools (Claude Code, Gemini CLI, OpenAI Codex, Copilot CLI, etc.).
- When you remove a provider, MCPX will delete the corresponding provider config file. Global providers (like Kimi) are cleaned up accordingly.
- You can customize per-server environment variables in the mcpServers entries; use the env field to specify required variables for each server.
- If you run into import or generation issues, run mcpx status to inspect sync status and mcpx sync to re-run generation.
- Ensure the CLI has network access if you’re importing or generating configs from online providers.
Related MCP Servers
gemini -tool
MCP server that enables AI assistants to interact with Google Gemini CLI, leveraging Gemini's massive token window for large file analysis and codebase understanding
better-chatbot
Just a Better Chatbot. Powered by Agent & MCP & Workflows.
ClaudeR
Connect RStudio to Claude, Codex, Gemini, and other AI assistants via MCP. Multi-agent orchestration, async execution, and zero-config setup with uvx.
forge-orchestrator
Forge Orchestrator: Multi-AI task coordination. File locking, knowledge capture, drift detection. Rust.
mcp-config-manager
Manage MCP server configs across Claude, Gemini & other AI systems. Interactive CLI for server enable/disable, preset management & config sync.
gemini-webapi
MCP server for Google Gemini — free image generation, editing & chat via browser cookies. No API keys needed.