Local Pm

🚀 Local PM

A lightweight, self-hosted project management tool. It comes with a built-in MCP (Model Context Protocol) server, allowing AI assistants to directly manage your projects, tickets, and teams.

🚀 Quick Start

Local PM is a powerful project management tool that integrates an MCP server, enabling seamless interaction with AI assistants. You can start using it by following the installation steps below.

✨ Features

• Kanban Board: Drag-and-drop ticket management with columns for Todo, In Progress, and Done.
• Projects: Organize work using customizable projects with icons, colors, and prefixes.
• Teams: Assign tickets to teams for better organization.
• Tickets: Full-featured tickets with priority levels (Urgent, High, Medium, Low), due dates, custom colored labels, subtasks with completion tracking, ticket dependencies, and rich text descriptions.
• MCP Server: AI-native project management via the Model Context Protocol.
• Self-Hosted: Your data remains on your machine.
• Docker Ready: One-command deployment.

📦 Installation

Using Docker (Recommended)

1. Clone the repository:

git clone https://github.com/your-username/local-pm.git
cd local-pm

2. Start the containers:

docker-compose up -d

3. Access the app at http://localhost:3010

Manual Installation

1. Clone the repository and install dependencies:

git clone https://github.com/your-username/local-pm.git
cd local-pm
npm install

2. Set up environment variables:

cp .env.example .env
# Edit .env with your MongoDB connection string

3. Run the development server:

npm run dev

📚 Documentation

MCP Server Setup

The MCP (Model Context Protocol) server allows AI assistants like Claude to interact directly with your project management data.

Building the MCP Server

cd mcp-server
npm install
npm run build

Adding to Claude Code (Global)

claude mcp add --scope user local-pm node "/path/to/local-pm/mcp-server/dist/index.js"

Adding to Claude Desktop

Add to your Claude Desktop config (~/.claude/claude_desktop_config.json):

{
  "mcpServers": {
    "local-pm": {
      "command": "node",
      "args": ["/path/to/local-pm/mcp-server/dist/index.js"],
      "env": {
        "LOCAL_PM_URL": "http://localhost:3010"
      }
    }
  }
}

MCP Tools Reference

The MCP server exposes 18 tools for comprehensive project management:

Project Tools

Team Tools

Ticket Tools

Board & Subtask Tools

🔧 Technical Details

How MCP Enhances AI Development

What is MCP?

Model Context Protocol (MCP) is an open standard that enables AI assistants to interact with external tools and data sources. Instead of just chatting, AI can take actions in the real world through well-defined tool interfaces.

Benefits for AI-Assisted Development

1. Persistent Task Tracking: AI can create tickets for features it's implementing, track progress across coding sessions, and never lose context on what was done or what's pending.
2. Structured Workflow: AI breaks down complex features into subtasks, sets priorities and due dates, and manages dependencies between tickets.
3. Project Organization: AI can organize work into logical projects, assign tasks to teams, and maintain a clear overview of all work.
4. Seamless Integration: It works directly in your AI coding workflow, eliminating the need for context switching to external tools. AI reads and updates tickets as it works.

Example Workflow

You: "Create a project for our new authentication system"
AI: [Creates project AUTH with relevant description]
You: "Break down the login feature into tickets"
AI: [Creates tickets for:
  - AUTH-1: Implement login form UI
  - AUTH-2: Create authentication API endpoint
  - AUTH-3: Add JWT token handling
  - AUTH-4: Implement session management
  Sets AUTH-2 as blocking AUTH-3 and AUTH-4]
You: "Start working on the login form"
AI: [Moves AUTH-1 to In Progress, implements the feature,
     then moves to Done when complete]

Why Local & Self-Hosted?

• Privacy: Your project data stays on your machine.
• Speed: No network latency for AI tool calls.
• Control: Customize and extend as needed.
• Offline: Works without an internet connection.

Tech Stack

• Frontend: Next.js 15, React, Tailwind CSS
• Backend: Payload CMS 3.0
• Database: MongoDB
• MCP Server: TypeScript, @modelcontextprotocol/sdk

📄 License

This project is built with Payload CMS.