scottiesan/backlogr-mcp

Backlogr MCP Server

Backlogr is a Model Context Protocol (MCP) server designed to bridge the gap between AI development assistants (like ChatGPT and Claude) and project management workflows. It provides a standardized interface for LLMs to plan, create, and track development projects through a structured backlog system.

Core Capabilities

🎯 AI-Powered Project Planning

• Project Management: Create, list, fetch, and delete development projects
• Task Orchestration: Full CRUD operations for tasks with prioritization (P0-P3) and status tracking
• Requirements Management: Define and link functional requirements to development tasks
• Smart Search: Natural language search across projects, tasks, and requirements

🤖 MCP Integration for Coding Agents

• Standardized Interface: MCP-compliant tools that work with any MCP-enabled environment
• Task Assignment: Coding agents can pull development tasks directly from the backlog
• Context Awareness: Provides rich context about requirements and project scope
• Real-time Updates: Agents can update task status and progress as they work

🔧 Technical Architecture

• Built on MCP SDK: Leverages the official Model Context Protocol SDK for compatibility
• Supabase Backend: Scalable PostgreSQL database with real-time capabilities
• TypeScript Foundation: Type-safe development with comprehensive error handling
• RESTful API: Additional HTTP endpoints for web integration

Key Features

For Project Managers & LLMs:

• Project Creation: Start new development initiatives with structured backlogs
• Task Prioritization: Organize work with clear priority levels (P0-P3)
• Status Tracking: Monitor progress through Open → In Progress → Review → Done lifecycle
• Requirement Linking: Connect business requirements to implementation tasks

For Development Agents:

• Task Discovery: Search and fetch available development work
• Context Retrieval: Access complete project context including linked requirements
• Progress Reporting: Update task status and provide completion updates
• Collaboration: Work within team projects with proper access controls

Use Cases

1. LLM-Assisted Planning: ChatGPT/Claude can help structure entire projects by creating epics, stories, and tasks
2. Automated Development: Coding agents pull tasks, implement features, and mark them complete
3. Team Coordination: Multiple agents can work on the same project without conflicts
4. Progress Monitoring: Real-time visibility into what's being worked on and completed

Integration Ready

• MCP Standard: Works with any MCP client (Anthropic Claude, Cursor, etc.)
• Supabase Powered: Enterprise-grade database with authentication and real-time features
• Extensible: Easy to add new tools and capabilities as the MCP standard evolves
• Self-Hosted: Full control over your data and infrastructure

Quick Start

1. Install Dependencies:

   npm install
   

2. Configure Environment: Copy .env.example to .env and set your Supabase credentials:

   cp .env.example .env
   

3. Start Development Server:

   npm run dev
   

4. Connect to MCP Client: Configure your MCP client (Claude, Cursor, etc.) to connect to the running server.

API Endpoints

The server provides both MCP tools and RESTful endpoints:

• MCP Tools: search, fetch, backlogr_project, backlogr_task, backlogr_requirement
• HTTP API: /messages, /rpc, /search, /fetch endpoints

Database Schema

The app uses a Supabase PostgreSQL database with tables for:

• projects - Development projects
• tasks - Individual work items with priorities and status
• requirements - Functional requirements
• project_members - Team member access control
• task_requirements - Linking table between tasks and requirements

Development

# Build for production
npm run build

# Start development server with hot reload
npm run dev

License

ISC License - see package.json for details