Spec Workflow MCP Server

@kingkongshot

リポジトリ

基本情報

カテゴリ

推論

トランスポート

stdio

作者

kingkongshot

投稿者

link

Spec Workflow MCP Server について

Guide AI to systematically complete software development through a structured **Requirements → Design → Tasks** workflow, ensuring code implementation stays aligned with business needs.

#spec#kiro#workflow

Spec Workflow MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "spec-workflow": {
      "command": "npx",
      "args": [
        "-y",
        "spec-workflow-mcp@latest"
      ]
    }
  }
}

Spec Workflow MCP Server のツール

Manage intelligent writing workflow for software project requirements, design, and task documents. Supports initialization, checking, skipping, confirmation, and task completion operations (single or batch).

概要

What is Spec Workflow Mcp?

Spec Workflow Mcp is an MCP server that helps developers create structured specification documents (requirements, design, tasks) using AI assistants. It integrates with any MCP-compatible client and is designed for anyone building software projects.

How to use Spec Workflow Mcp?

Install the server via npx (npx -y spec-workflow-mcp@latest) and add it to your MCP client’s configuration (e.g., Claude Desktop, Cursor, or Windsurf). Then ask your AI to “use spec workflow” for a project—the AI will automatically detect project status and guide you through creating requirement, design, and task documents step by step.

Key features of Spec Workflow Mcp?

Creates structured spec documents (requirements, design, tasks)
Supports single-module and multi-module project organization
Automatic progress tracking with confirmation files
Works with Claude Desktop, Cursor, Windsurf, Zed, and more
Requires Node.js ≥ v18.0.0
Step-by-step review with user confirmation before proceeding

Use cases of Spec Workflow Mcp?

Starting a new project – quickly generate requirements, design, and tasks from a high-level description
Continuing an existing project – automatically detect progress and pick up where you left off
Building a user authentication system – create detailed specs for a specific module
Organizing multi-module projects – keep specs for different modules in separate subdirectories

よくある質問

Which MCP clients are supported?

Claude Code, Claude Desktop, Cursor, Cline (VS Code extension), Windsurf (Codeium), VS Code with an MCP extension, and Zed.

What are the runtime requirements?

Node.js v18.0.0 or higher and npm or yarn.

How do I install Spec Workflow Mcp?

The recommended method is `npx -y spec-workflow-mcp@latest`. You can also install from source by cloning the repository, running `npm install && npm run build`, and pointing your client to the built `index.js`.

How can I check the current progress of a project?

Use the `specs-workflow` tool with `action.type="check"` and `path` set to your specs directory. The server will return the current status.

How do I mark a task as complete?

Use the `specs-workflow` tool with `action.type="complete_task"` and `taskNumber` set to the current task number. The server will automatically return the next pending task.

「推論」の他のコンテンツ

quarkus-langchain4j-workshop

quarkusio

Quarkus LangChain4J Workshop that demonstrates both single AI service capabilities and Agentic AI orchestration

Node Code Sandbox MCP 🛠️

mozicim

# 🐢🚀 Node.js Sandbox MCP ServerThis repository hosts a Node.js server that implements the Model Context Protocol (MCP) for running JavaScript in isolated Docker containers. It allows for on-the-fly npm dependency installation, making it easy to execute code safely and efficient

Part 1. Real-Time LangGraph Agent with MCP Tool Execution

junfanz1

This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independent

Proplan Mcp

King-Proplan

MCP server that gives Claude persistent project memory — roadmap, session history, and codebase context. Type continue and Claude picks up exactly where you left off

MCP Advanced Reasoning Server for Cursor AI

AzDeltaQQ

This is a very basic implementation of an Mcp-Reasoning-Server for Cursor AI .