MCP Assistant Server

🚀 MCP Assistant

MCP Assistant is a project designed to offer developers core functions such as task analysis, tool recommendation, and context management.

🚀 Quick Start

📦 Installation

npm install

🚀 Launch the Service

npm run dev

✨ Features

Core Functions

• Task Analyzer: Analyze the task description input by the user and extract key information.
• Tool Recommendation System: Intelligently recommend tools based on task characteristics.
• Context Manager: Maintain context information during task execution.

Technical Architecture

src/
├── controllers/     # Controller layer
│   └── LLMApiController.ts
├── core/           # Core function implementation
│   ├── TaskAnalyzer.ts
│   ├── ToolRecommender.ts
│   └── ContextManager.ts
├── services/       # Service layer
│   ├── LLMService.ts
│   └── MCPDiscoveryService.ts
├── routes/         # Route definition
│   └── llmApiRoutes.ts
└── types/          # Type definition
    └── interfaces.ts

💻 Usage Examples

Usage in Cline

Modify the Configuration File

Create or modify the settings.json file in the root directory of your project and add the following content:

{
  "mcpServers": {
    "assistant": {
      "command": "node",
      "args": ["./path/to/mcp-assistant-server/build/index.js"],
      "env": {
        "PORT": "3000",
        "LOG_LEVEL": "info"
      }
    }
  }
}

How to Use

In Cline, you can use the following commands through the command panel (Cmd/Ctrl + Shift + P):

1. MCP: Analyze Task - Analyze the currently selected code or text.
2. MCP: Recommend Tools - Get tool recommendations for the current task.
3. MCP: Update Context - Update task context information.

Usage in Cursor

Modify the Configuration File

Create or modify the cursor/settings.json file in the root directory of your project and add the following content:

{
  "mcpServers": {
    "assistant": {
      "command": "node",
      "args": ["./path/to/mcp-assistant-server/build/index.js"],
      "env": {
        "PORT": "3000",
        "LOG_LEVEL": "info"
      }
    }
  }
}

How to Use

In Cursor, you can use the following commands through the command panel (Cmd/Ctrl + Shift + P):

1. MCP: Analyze Task - Analyze the currently selected code or text.
2. MCP: Recommend Tools - Get tool recommendations for the current task.
3. MCP: Update Context - Update task context information.

You can also use it in plugins through the Cursor API:

import { workspace } from 'cursor';

// Analyze the task
const analysis = await workspace.mcp.callTool('assistant', 'analyze_task', {
  description: 'Current task description',
  context: {
    // Context information
  }
});

// Get tool recommendations
const recommendations = await workspace.mcp.callTool('assistant', 'recommend_tools', {
  taskId: 'current-task-id'
});

🔧 Technical Details

Task Analyzer

Analyze the task description input by the user and extract key information:

• Task Type Recognition
• Parameter Extraction
• Context Association
• Dependency Analysis

Tool Recommendation System

Intelligently recommend tools based on task characteristics:

• Rule-based Recommendation
• Hybrid Recommendation Algorithm
• Context Awareness
• Historical Data Analysis

Context Manager

Maintain context information during task execution:

• Status Tracking
• Data Persistence
• Session Management
• Error Recovery

📚 Documentation

API Documentation

• LLM API: Provide an interface for interacting with large language models.
• Task Analysis: Analyze the task description and return a structured result.
• Tool Recommendations: Recommend tools based on task requirements.

Configuration Options

• PORT: The port number on which the service runs, with a default of 3000.
• LOG_LEVEL: Log levels, supporting debug, info, warning, error.

Project Structure

src/
├── controllers/     # Controller layer
│   └── LLMApiController.ts
├── core/           # Core function implementation
│   ├── TaskAnalyzer.ts
│   ├── ToolRecommender.ts
│   └── ContextManager.ts
├── services/       # Service layer
│   ├── LLMService.ts
│   └── MCPDiscoveryService.ts
├── routes/         # Route definition
│   └── llmApiRoutes.ts
└── types/          # Type definition
    └── interfaces.ts

🤝 Contribution Guidelines

Contributions are welcome! Please refer to the CONTRIBUTING.md file to learn how to participate in the development.

📄 License

This project is licensed under the MIT License.