mcp -generator

The Manifest MCP Server Generator is designed to help you rapidly create MCP servers that expose data sources to AI agents with built-in access control and governance. It focuses on generating the plumbing and scaffolding needed to connect a datasource to MCP-based tools, enabling quick experimentation and iteration without building every component from scratch. Once generated, you get a runnable MCP server with prompts and tooling that streamline querying, data access, and basic resource creation for downstream tasks.

Prerequisites:

• Node.js (version 14+ recommended) and npm/yarn installed on your machine
• Internet connection to fetch the generator package from npm

Installation steps:

1. Ensure Node.js and npm are installed:
   • macOS/Linux: curl -fsSL https://get.npmjs.org/install.sh | sh
   • Windows: install Node.js from https://nodejs.org/
2. Install or run the generator via npx (no global install required):
   • npm: npx -y mnfst-mcp-server-generator
   • or with npm/yarn if you prefer: npm install -g mnfst-mcp-server-generator mnfst-mcp-server-generator
3. Follow the interactive prompts (if any) to configure your MCP server, or customize the generated scaffold as needed.
4. Start the generated server per the output instructions (typically node-based or Docker-ready).

Notes:

• This project is a Proof-Of-Concept and may not be production-ready. Validate access controls and data exposure before using in a live environment.
• If you encounter permission or network issues during npx execution, try clearing npm cache or using a different registry.
• The generator focuses on quick scaffolding; you can customize prompts, tools, and resources after generation.
• If the generator outputs a Docker-ready container, you can containerize the server for consistent deployments.

Environment Variables (examples you may need to set):

• MCP_API_BASE_URL: Base URL for MCP API endpoints
• MCP_AUTH_TOKEN: Token used for authenticating with MCP services
• DATA_SOURCE_CONNECTION_STRING: Connection string to your datasource
• LOG_LEVEL: Logging verbosity (e.g., info, debug, warn, error)

Common issues:

• Module not found errors after generation: ensure dependencies are installed (npm install) in the generated project
• Access control not enforcing as expected: review the generated governance/configuration files and adjust policies accordingly