Disler Aider MCP Server

🚀 Aider's AI Model Interface: MCP Server

This project is a complete implementation of an MCP (Modular Component Protocol) server, specifically designed for Aider's AI models. The server offers two core functions:

1. AI Coding Tool: Allows intelligent modification of files using a specified AI model and returns the change results.
2. Model List Tool: Lists all supported AI models.

✨ Features

• MCP Compatibility: Fully adheres to the MCP protocol, ensuring seamless interaction with clients and consoles.
• Extensible Plugin Architecture: The atomic layer design simplifies the addition of new features or modification of existing ones.
• High Reliability: All components have undergone rigorous testing to ensure server stability and availability.

📦 Installation

Project Structure

.
├── ai_docs                   # Directory for AI-related documents
│   ├── just-prompt-example-mcp-server.xml
│   └── programmable-aider-documentation.md
├── pyproject.toml            # Project metadata and dependency management
├── README.md                 # Project documentation file
├── specs                     # Normative documents
│   └── init-aider-mcp-exp.md
├── src                       # Source code directory
│   └── aider_mcp_server      # Main server code package
│       ├── __init__.py       # Package initialization file
│       ├── __main__.py       # Server startup entry point
│       ├── atoms             # Core, reusable components (pure functions)
│       │   ├── __init__.py
│       │   ├── data_types.py # Pydantic model defining data structures
│       │   ├── logging.py    # Custom logging settings
│       │   ├── tools         # Specific implementation of each tool
│       │   │   ├── __init__.py
│       │   │   ├── aider_ai_code.py # Logic for the AI coding tool
│       │   │   └── aider_list_models.py # Logic for the model list tool
│       │   └── utils.py      # Utility functions and constants (e.g., default model)
│       ├── server.py         # MCP server logic, registering tools, processing requests, etc.
│       └── tests             # Unit tests and integration tests
│           ├── __init__.py
│           └── atoms         # Testing atomic layer components
│               ├── __init__.py
│               ├── test_logging.py # Testing logging functionality
│               └── tools     # Testing tool logic
│                   ├── __init__.py
│                   ├── test_aider_ai_code.py # Testing the AI coding tool
│                   └── test_aider_list_models.py # Testing the model list tool

Installation Steps

1. Clone the repository:

   git clone [your-repository-url]
   cd aider-mcp-server
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Start the server:

   python -m aider_mcp_server.__main__ --editor-model <model-name>
   

🔧 Technical Details

Core Components

1. server.py

• Initialization: Loads configuration files and sets up logging.
• Tool Registration: Loads all available tools from atoms/tools.
• Request Handling: Parses client requests, routes them to the corresponding tool logic, and returns results according to the MCP protocol.

2. __main__.py

• Provides a command-line interface (CLI), parses startup parameters such as --editor-model, and initializes server components.

Tool Implementations

AI Coding Tool

Functionality

• Receives a file path and modification instructions.
• Generates modified content using the specified AI model.
• Returns the change results and comparison information with the original file.

Example Request

{
  "name": "aider_ai_code",
  "parameters": {
    "file_path": "src/file.py",
    "change_description": "Add logging statements",
    "model_name": "openai/gpt-4"
  }
}

Return Result

{
  "success": true,
  "diff": "--- a/src/file.py\n+++ b/src/file.py\n@@ -1,5 +1,6 @@\n import sys\n+import logging\n \n def main():
     print(\"Hello World!\")
+    logging.info(\"Started the program.\")
 }

Model List Tool

Functionality

• Filters supported AI models based on a substring.
• Returns a list of matching models.

Example Request

{
  "name": "aider_list_models",
  "parameters": {
    "query": "gpt"
  }
}

Return Result

[
  "openai/gpt-3.5-turbo",
  "anthropic/claude-2",
  "google/palm-1"
]

Logging System

Level Classification

• DEBUG: Detailed debugging information.
• INFO: Regular operation logs.
• WARNING: Warnings for potential issues.
• ERROR: Records of serious errors.
• CRITICAL: Critical events such as system crashes.

Output Format

[asctime] - [levelname] - [filename] - [message]

Testing Framework

Unit Tests

Ensure that the behavior of each tool and component meets expectations.

Integration Tests

Verify the correctness of the overall server functionality, including the interaction and collaboration between different tools.

Extension Instructions

Developing New Tools

1. Create a new tool class in the atoms/tools directory.
2. Implement the Tool interface and register it in server.py.

Supporting New Models

1. Add a new AI model adapter.
2. Update the model list in the aider_list_models tool.

📚 Documentation

Project Characteristics

• Modular: Each component is developed independently, facilitating maintenance and expansion.
• Pluggable: Supports dynamic loading of new features without modifying the main program.
• High Availability: Rigorous testing ensures stable server operation.

Use Cases

Developers

• Quickly integrate AI capabilities into the development environment.
• Flexibly select different AI models for experimentation and comparison.

Enterprise Applications

• Integrate into the CI/CD pipeline to automatically perform code reviews and optimizations.
• Integrate into internal toolchains to improve development efficiency.