ContextEngine MCP Server

A Model Context Protocol (MCP) server that provides an interface to the ContextEngine system - a comprehensive Documentation Driven Development (DDD) methodology. This MCP server enables AI assistants to access ContextEngine's structured workflows, context management tools, and knowledge graph system for documentation-first development processes, eliminating repetitive context provision and ensuring code generation follows documented requirements.

🚀 What is ContextEngine?

ContextEngine is an MCP server that implements a comprehensive Documentation Driven Development (DDD) system. It provides structured workflows, context management, and tools that enable AI assistants and humans to collaborate effectively through documentation-first development methodologies.

Core Philosophy

ContextEngine addresses the fundamental challenges of AI-human collaboration in software development:

• Context Management: Eliminates repetitive context provision by creating persistent documentation repositories

• Quality Assurance: Ensures code generation follows documented requirements and specifications

• Scalability: Handles 3x to 30x more information per person than traditional development methods

• Business Alignment: Maintains clear connection between technical implementation and business value

Related MCP server: Code Research MCP Server

Features

• Documentation-Driven Workflows: 9 standardized workflows for different development activities

• Context Engineering: Dynamic context composition and provision through knowledge graph

• Structured Documentation: Plan and task documents with hierarchical organization

• Local Documentation Setup: Automatic creation of organized folder structure and configuration files

• TypeScript: Full type safety and modern development experience

• Authentication: Built-in support for API keys and authentication

• Comprehensive Logging: Structured logging with proper MCP compatibility

🛠️ Installation

Requirements

• Node.js >= v18.0.0

• Cursor, Claude Code, VSCode, Windsurf or another MCP Client

Connecting to MCP Clients

"mcp": {
  "servers": {
    "context-engine": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "context-engine", "--api-key", "YOUR_API_KEY"]
    }
  }
}

🔨 Available Tools

ContextEngine provides tools for DDD workflow execution:

1. start_context_engine

Starts the ContextEngine system and automatically sets up the local documentation structure. This tool:

• Initializes the ContextEngine via API call to establish system awareness

• Creates local documentation structure with organized folders for DDD workflows

• Sets up configuration files with default settings and workflow definitions

• Provides comprehensive feedback about both remote and local setup status

The tool creates a .context-engine directory structure:

.context-engine/
├── implementation/       # For completed documentation and implementations
├── requirements/         # For requirements and specifications  
└── config/              # For configuration files
    ├── settings.json     # ContextEngine settings
    └── workflows.json    # Workflow configurations

Response Format: Returns a combined status showing both API response and local setup results with clear emoji indicators:

• 📁 Success: Documentation structure setup completed

• ⚠️ Warning: Setup completed with issues

• ❌ Error: Setup failed (but API call succeeded)

CLI Arguments

Your MCP server accepts the following CLI flags:

• --transport <stdio|http> – Transport to use (stdio by default)

• --port <number> – Port to listen on when using http transport (default 3000)

• --api-key <key> – API key for authentication (if needed)

• --server-url <url> – Custom server URL (defaults to https://contextengine.in)

📚 Usage

1. Start ContextEngine: Use the start_context_engine tool to initialize the DDD system and set up local documentation structure and local documentation setup results

2. Select Workflow: Choose from 9 available workflows based on your development objective

3. Execute Workflow: Follow the structured workflow phases to complete your development task

4. Integrate with AI assistants: Connect to Cursor, VS Code, Claude Code, etc.

📄 License

MIT