Roo-Code -installer
An MCP server installer for Roo Code. Handles the installation from Github, requirements, package management, update, installation and MCP Json configuration.
claude mcp add --transport stdio robertheadley-roo-code-mcp-installer python roo.py
How to use
Roo MCP Installer is a command-line utility that streamlines the installation and management of Roo MCP servers. It fetches MCP server code directly from Git repositories (including GitHub URLs or user/repo slugs), detects the project type (Node.js, Python, Go, Rust, etc.), installs dependencies, and registers the MCP within the Roo environment. The tool supports installation scopes (global or project-specific), subdirectory installations, automatic environment variable configuration, and optional demo or debug modes to preview actions without making changes. Use it to install new MCPs from a repository by supplying a repo input (URL, slug, or slug with a subdirectory), choose a scope, and optionally skip interactive env prompts or run in demo mode to see what would happen.
How to install
Prerequisites:
- Python 3.11 or higher
- Git installed and available on your PATH
- Access to the internet
- (Optional) Node.js, Go, Rust toolchains if the MCP requires building dependencies
Install steps:
-
Clone or download the Roo MCP Installer repository and navigate into it. git clone https://github.com/<your-org>/rocmcp-installer.git cd rocmcp-installer
-
Create and activate a Python virtual environment (recommended): python -m venv venv source venv/bin/activate # on Unix/macOS venv\Scripts\activate # on Windows
-
Install Python dependencies listed in requirements.txt: python -m pip install -r requirements.txt
-
Run the installer helper script to install an MCP from a repository (example): python roo.py install username/my-cool-mcp
-
Optional: run in demo mode to preview actions without changing the filesystem: python roo.py install username/my-cool-mcp --demo
-
If you want global installation of MCPs (default behavior in Roo), ensure the installer has the appropriate permissions and follow the prompts to set the scope to global. To install for a specific project, pass --scope project and operate within the project directory.
Notes:
- Ensure you have network access to fetch repositories and install dependencies.
- If a repository contains a subdirectory (e.g., username/repo:subpath), the installer will target that subdirectory for installation and configuration.
- You may need to install language-specific toolchains (Node.js/npm, Go, Rust/Cargo) if the MCP requires building or dependency resolution beyond Python.
Additional notes
Tips and caveats:
- Environment variable handling is interactive by default, automatically detecting variables from mcp.json, README.md, or .env.example. You can skip prompts with --skip-env. Sensitive values are masked during input.
- The installer lists installed MCPs for both global and project scopes and provides log viewing via the logs command. Use --lines or --follow to tail logs in real time.
- If you encounter missing tool errors, the installer can suggest automatic global installations where possible (e.g., npm install -g bun). Ensure your PATH includes the tools the MCP may need.
- For complex MCPs, pre-reading the repository’s README and mcp.json (if present) helps anticipate required environment variables and build steps.
- The demo mode is helpful for checking the user interface and flow without performing actual changes.
Related MCP Servers
context7
Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
web-eval-agent
An MCP server that autonomously evaluates web applications.
mcp -spec-driven-development
Spec-Driven Development MCP Server, not just Vibe Coding
frontmcp
TypeScript-first framework for the Model Context Protocol (MCP). You write clean, typed code; FrontMCP handles the protocol, transport, DI, session/auth, and execution flow.
editor
MCP Server for AI automation of the PlayCanvas Editor
pluggedin-app
The Crossroads for AI Data Exchanges. A unified, self-hostable web interface for discovering, configuring, and managing Model Context Protocol (MCP) servers—bringing together AI tools, workspaces, prompts, and logs from multiple MCP sources (Claude, Cursor, etc.) under one roof.