mcp-gearbox-cli

MCP Gearbox CLI is a Python-based tool that helps you discover, download, and configure MCP (Model Context Protocol) servers for AI agents. It provides an interactive workflow to select an AI agent, fetch the appropriate MCP server, and prepare it for deployment across supported platforms. Once configured, you can deploy and manage MCP servers that integrate with your preferred agent runtimes. The CLI focuses on streamlining the process of obtaining compatible MCP servers and applying sensible defaults for cross-platform setups, reducing manual setup time for developers and teams.

Using the gearbox, you can search for available MCP servers, initialize configuration globally or per-project, and list or remove configured servers. The tool supports a variety of AI agents such as Copilot, Continue, Kiro, Cursor, Claude Code, Gemini CLI, LM Studio, and Antigravity, among others. This makes it easier to stay aligned with the ecosystem and ensure your environment is prepared to run MCP-enabled agents with minimal boilerplate.

Prerequisites:

• Python 3.8+ and pip installed on your system
• Optional: uv (recommended) or uvx for installation and execution workflows
• Git (if you plan to clone repositories)

Installation steps (preferred via uv):

1. Install the Gearbox CLI from the repository:

uv tool install mcp-gearbox-cli --from git+https://github.com/rohitsoni007/mcp-gearbox-cli

2. If you prefer direct Python-based usage, install via pip (as an alternative):

pip install mcp-gearbox

3. Global installation via npm (if you’re aligning with npm tooling):

npm install -g mcp-gearbox

4. Clone the repository locally and run the CLI (development mode):

git clone https://github.com/rohitsoni007/mcp-gearbox-cli
cd mcp-gearbox-cli
uv sync

Prerequisites recap:

• Python 3.8+ (for the Python-based CLI)
• uv/uvx or pip/npm as preferred installation method
• Git for cloning during development

Note: The Gearbox CLI is designed to be cross-platform and works on Windows, macOS, and Linux.

Tips and common considerations:

• The MCP Gearbox CLI focuses on automating the download and configuration of MCP servers. When configuring servers, you may need to tailor environment variables and paths to your project structure.
• If you encounter issues with module discovery, ensure your PYTHONPATH includes the directory where the gearbox CLI modules reside, or run via the -m invocation as shown in the mcp_config example.
• For large MCP server downloads, consider enabling verbose logging to diagnose network or integrity issues by setting MCP_GEARBOX_LOG_LEVEL=DEBUG.
• Always verify compatibility with your AI agent version before linking a server to an agent in your mcp configuration.
• If you’re using uv/uvx, you can follow the installation steps in the README to keep your environment consistent with the Gearbox CLI.