MCP Vibecoder

🚀 Vibe-Coder MCP Server

A Model Context Protocol (MCP) server for LLM-based coding that implements a structured development workflow.

🚀 Quick Start

This MCP server helps LLMs build features by providing the following capabilities:

• A structured feature clarification process with guided questions
• Generation of product requirement documents (PRDs) and implementation plans
• Phased development with task tracking
• Progress tracking and status reporting
• Document storage and retrieval capabilities

✨ Features

Resources

• Feature details, PRDs, and implementation plans
• Progress reports and status tracking
• Phase and task details

Tools

• start_feature_clarification - Initiate the feature clarification process
• provide_clarification - Answer clarification questions about features
• generate_prd - Generate product requirement documents and implementation plans
• create_phase - Create a development phase for a feature
• add_task - Add tasks to a development phase
• update_phase_status - Update the status of a phase
• update_task_status - Update the completion status of a task
• get_next_phase_action - Get guidance on the next action
• get_document_path - Get the path of the generated document
• save_document - Save a document to a specific location

Prompts

• feature-planning - A prompt template for planning feature development

📚 Documentation

Document Storage

The server includes a hybrid document storage system with the following features:

1. Automatically save generated documents (PRDs, implementation plans) to files.
2. Maintain an in-memory copy for quick access.
3. Allow clients to retrieve document paths and save them to custom locations.

Default Storage Location

Documents are stored in the documents/{featureId}/ directory by default, and the file names are based on the document type:

• documents/{featureId}/prd.md - Product requirement document
• documents/{featureId}/implementation-plan.md - Implementation plan

Custom Storage

You can use the save_document tool to save documents to a custom location:

{
  "featureId": "feature-123",
  "documentType": "prd",
  "filePath": "/custom/path/feature-123-prd.md"
}

Path Retrieval

To get the path of a document, use the get_document_path tool:

{
  "featureId": "feature-123",
  "documentType": "prd"
}

This will return the path and whether the document has been saved to disk.

📦 Installation

Development

Install dependencies:

npm install

Build the server:

npm run build

Develop with automatic rebuild:

npm run watch

Usage with Compatible MCP Clients

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "vibe-coder-mcp": {
      "command": "/path/to/vibe-coder-mcp/build/mcp-server.js"
    }
  }
}

Debugging

Since the MCP server communicates via standard input and output, debugging can be challenging. We recommend using MCP Inspector, which is provided as a package script:

npm run inspector

The Inspector will provide a URL to access the debugging tools. Open it in your browser.

🔧 Technical Details

The server is implemented using the advanced McpServer class from the Model Context Protocol TypeScript SDK, which simplifies the process of creating an MCP server and provides a clean API to define resources, tools, and prompts.

import { McpServer, ResourceTemplate } from "@modelcontextprotocol/sdk/server/mcp.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";

// Create an MCP server
const server = new McpServer({
  name: "Vibe-Coder",
  version: "0.3.0"
});

// Add resources
server.resource(
  "features-list",
  "features://list",
  (request) => {
    // Logic to handle the request
  }
);

// Add tools
server.tool(
  "generate_prd",
  (request) => {
    // Logic to generate a PRD
  }
);

// Add prompts
server.prompt(
  "feature-planning",
  (request) => {
    // Logic to handle the prompt
  }
);

In conclusion, by following the above steps, you can set up a fully functional MCP server to assist in LLM-based coding development.