Agile AI Workflow MCP Server

This project is a fully MCP-compatible VSCode AI workflow server, supporting agile team multi-role collaboration (BA, TL, DEV, QA).

Features

• MCP server with automatic discovery and invocation of multiple "role tools"
• Role tools include:
  • BA (Business Analyst): Generate user stories (YAML)
  • TL (Tech Lead): Split user stories into technical tasks (YAML)
  • DEV (Developer): Generate pseudocode and implementation suggestions (Markdown)
  • QA (Tester): Generate test cases (YAML)
• Each tool is an independent MCP Tool, supporting concurrent tasks, status, and result queries
• Supports OpenAI GPT-4o (default, requires API Key configuration)
• Supports local Ollama (optional, see askLLM.ts)

Quick Start

1. Install dependencies

npm install

2. Configure OpenAI Key

Create a .env file in the root directory:

OPENAI_API_KEY=sk-your-real-openai-key

3. Start the server

npm start

Project Structure

.
├── tools/         # Implementation of each agile role tool
├── prompts/       # LLM prompts for each role
├── utils/askLLM.ts# LLM call wrapper (supports OpenAI/Ollama)
├── taskManager.ts # Task status and result cache
├── server.ts      # Main server entry
├── manifest.json  # MCP plugin metadata
├── .env           # OpenAI key (ignored by .gitignore)
└── .gitignore

Notes

• Uses OpenAI by default. To use local Ollama, modify askLLM.ts
• Compatible with VSCode MCP plugin auto-discovery and invocation
• For learning and agile team automation reference only

Alternative Start (Stdio Mode)

To run the server in stdio mode (for VSCode MCP extension integration):

npm run --prefix /Users/$(whoami)/Desktop/agile-ai-workflow start:stdio