PRD

Flagship Model Context Protocol server for generating Product Requirement Documents (PRDs) from codebase context.

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio saml1211-prd-mcp-server npx -y prd-creator-mcp \
  --env OPENAI_MODEL="default" \
  --env PRD_TEMPLATE="standard" \
  --env OPENAI_API_KEY="your-openai-api-key" \
  --env ANTHROPIC_MODEL="claude-1.3" \
  --env LOCAL_MODEL_PATH="/path/to/local-model" \
  --env ANTHROPIC_API_KEY="your-anthropic-api-key" \
  --env GOOGLE_GEMINI_API_KEY="your-google-gemini-api-key"

How to use

This MCP server is a specialized tool for creating Product Requirements Documents (PRDs) via the MCP protocol. It exposes a set of capabilities to generate PRDs from product descriptions, user stories, and requirements, using multiple AI providers or template-based generation as a fallback. You can request a complete PRD through the generate_prd tool, validate the resulting document with validate_prd, and manage templates and provider configurations at runtime using the protocol tools exposed by the MCP server. The server supports configuring providers (OpenAI, Google Gemini, Anthropic Claude, or local models), and it can gracefully fall back to templates if AI providers are unavailable. Integrations show how to connect the MCP server to clients like Claude Desktop, Glama.ai, Cursor, Roo Code, and Cline, enabling seamless MCP-based workflows.

How to install

Prerequisites:

Node.js v16 or higher
npm or yarn

Install from source:

Clone the repository:

git clone https://github.com/Saml1211/prd-mcp-server.git
cd prd-mcp-server

Install dependencies:

npm install

Build the project:

npm run build

Run locally:

npm start

For development with hot reload:

npm run dev

Alternative: Run via NPX (recommended in Quick Start):

npx -y prd-creator-mcp

Docker (optional):

docker pull saml1211/prd-creator-mcp
docker run -i --rm saml1211/prd-creator-mcp

Additional notes

Tips and notes:

Provide your API keys and model preferences through a .env file (see .env.example in the repo) or via the update_provider_config MCP tool at runtime. The server merges protocol/tool updates with environment variables, giving precedence to runtime changes.
The default PRD workflow generates PRDs using AI providers, with a template-based fallback when providers are unavailable. You can customize provider options (temperature, maxTokens, etc.) per request.
If integrating with external MCP clients, use the provided integration snippets for Claude Desktop, Glama.ai, Cursor, Roo Code, and Cline to add your server under a named MCP server entry (e.g., prd-creator).
Ensure you have the required environment variables set or provide credentials through the MCP tooling to avoid startup failures.
For troubleshooting, use health_check and get_logs endpoints to diagnose provider availability and recent activity.

Related MCP Servers

utcp-specification

289

The specification for the Universal Tool Calling Protocol

outline

111

It's an MCP server... for Outline (the documentation platform!)

CanvasMCPClient

Canvas MCP Client is an open-source, self-hostable dashboard application built around an infinite, zoomable, and pannable canvas. It provides a unified interface for interacting with multiple MCP (Model Context Protocol) servers through a flexible, widget-based system.

company-docs

AI-powered company knowledge MCP. Unified place for internal policies, values, documentation, and governance. Agents can search, cite, and answer questions using real company docs.

docmole

Dig through any documentation with AI - MCP server for Claude, Cursor, and other AI assistants

obsidian

MCP server for Obsidian vault management - enables Claude and other AI assistants to read, write, search, and organize your notes