Deskaid

🚀 Code Management Tool codemcp Usage Guide

This guide provides a comprehensive overview of codemcp, a code management tool designed to enhance development efficiency. It covers installation, configuration, usage, core features, advanced settings, troubleshooting, and more.

🚀 Quick Start

Prerequisites

• Install Python and pip if they are not already installed.
• Install codemcp using the following command:

pip install codemcp

Project Configuration

Create a codemcp.toml file in the root directory of your project with the following content:

[commands]
format = ["black ."]
test = ["pytest tests/ --cov"]

💻 Usage Examples

Project Initialization

Create a new project in Claude Desktop and add the initialization command:

Initialize codemcp with $PROJECT_DIR

Replace $PROJECT_DIR with the actual path to your project.

Interaction with Claude

Make code modifications through Claude and check the generated commits after each conversation:

git log --oneline --decorate

✨ Features

Automated Operations

• Code Formatting: Execute the !format command.
• Test Execution: Execute the !test command.

Commit History

Each conversation generates a commit. Use git reflog to view historical modifications.

📚 Documentation

Philosophy and Precautions

• Manual Intervention: Manually review the commit content after each conversation.
• Prevent Abuse: Avoid continuous long - term operations to prevent exceeding limits.
• Economy Mode: When the rate limit is reached, engage in other tasks such as code reviews or meetings.

Advanced Configuration

Project Prompts

Add an initial prompt in codemcp.toml:

project_prompt = """
Before beginning work on this feature, write a short haiku.  Do this only once.
"""

Custom Commands

Configure custom commands and their documentation:

[commands.test]
command = ["pytest tests/ --cov"]
doc = "Runs the test suite with coverage"

Troubleshooting

Development Mode

Run the debugging server:

PYTHONPATH=. mcp dev codemcp/__main__.py

Logging

Configure the global logging level:

[logger]
verbosity = "INFO"  # Options: DEBUG, INFO, WARNING, ERROR, CRITICAL

The log file is located at ~/.codemcp/codemcp.log.

Contribution Guide

Refer to CONTRIBUTING.md for contribution information.

Type Checking

The project uses pyright for type checking. Configure it as follows:

1. Type Stubs
   • Custom type stubs are stored in the stubs/ directory.
   • The configuration file mapping is in pyproject.toml.
2. Ignoring Strategies
   • Use specific ignoring strategies for files with complex dynamic types through tool.pyright.ignoreExtraErrors.

Execute the following command to ensure type checking passes:

./run_typecheck.sh

💡 Usage Tip

This tool is designed to improve development efficiency. It is recommended to combine it with manual reviews to ensure code quality.