mcp-claude-code

🥈Silver

MCP Claude Code enables Claude to directly execute instructions for modifying and improving project files. It benefits operations teams by automating code changes and improvements. It connects to Python-based workflows and integrates with MCP-compatible tools.

297360Updated 3d ago

Intermediate30min to implementautomation

Saves ~150 min per use

Quick InstallView Source

git clone https://github.com/SDGLBL/mcp-claude-code.git

Works with:

Claude

Overview

About This Skill

How to Use

1. **Prepare Your Environment**: Ensure your project uses a Python virtual environment (e.g., `venv` or `conda`). Install MCP-compatible tools like `mcp-server-python` and verify Claude Code is enabled in your MCP client. 2. **Run the Skill**: Use the prompt template above, replacing [TASK DESCRIPTION], [PROJECT_PATH], and [REQUIREMENTS] with your specific needs. For example: ``` Use MCP Claude Code to add a caching layer to the data processing pipeline in /src/pipeline/ using Redis. Ensure the cache invalidates after 5 minutes and includes error handling for Redis connection failures. ``` 3. **Review Changes**: After execution, review the generated summary and verify the changes in your IDE (e.g., VS Code, PyCharm). Check for any breaking changes or new dependencies. 4. **Test and Deploy**: Run your project's test suite (e.g., `pytest`) to ensure nothing is broken. Deploy changes incrementally, starting with a staging environment. 5. **Iterate**: If the changes introduce issues, use MCP Claude Code to roll back or refine the implementation. For example: ``` Use MCP Claude Code to revert the changes made to /src/api/legacy_client.py and restore the original synchronous implementation. ``` **Tips:** - Use specific file paths (e.g., `/src/` instead of `the project directory`) to avoid ambiguity. - Include constraints like `do not modify tests` or `preserve existing functionality` to guide the AI. - For large projects, break tasks into smaller chunks (e.g., refactor one module at a time).

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/SDGLBL/mcp-claude-code

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use MCP Claude Code to [TASK DESCRIPTION] in the [PROJECT_PATH] directory. Ensure all changes are [REQUIREMENTS, e.g., backward-compatible, follow PEP 8, include error handling]. Generate a summary of the changes made and any potential impacts. Example: 'Use MCP Claude Code to refactor the legacy API client in /src/api/ to use async/await and add comprehensive logging. Ensure the changes maintain backward compatibility and include unit tests.'

Example Output

### Summary of Changes

**Files Modified:**
1. `/src/api/legacy_client.py`
   - Refactored synchronous HTTP requests to use `aiohttp` for async/await support.
   - Added structured logging using `logging` module with log levels: DEBUG, INFO, WARNING, ERROR.
   - Introduced retry logic for transient failures (max 3 retries with exponential backoff).
   - Added type hints for all public methods and improved docstrings.

2. `/src/api/__init__.py`
   - Updated imports to include `aiohttp` and `logging_config` module.
   - Added `__all__` to expose only the public API surface.

3. `/tests/test_legacy_client.py`
   - Added async test cases for the new async methods using `pytest-asyncio`.
   - Included mock tests for retry logic and error handling.
   - Updated existing tests to work with the new async interface.

**Key Improvements:**
- **Performance:** Reduced average response time from 2.1s to 0.8s under load (tested with 100 concurrent requests).
- **Reliability:** Retry logic reduced transient failure rates from 5% to 0.2%.
- **Maintainability:** Type hints and docstrings make the codebase 40% easier to onboard new developers (based on internal feedback).

**Potential Impacts:**
- **Breaking Changes:** The synchronous `fetch_data()` method is now deprecated. Users should migrate to `fetch_data_async()`. A deprecation warning is logged when the old method is called.
- **Dependencies:** Added `aiohttp>=3.8.0` and `pytest-asyncio>=0.20.0` to `requirements.txt`.
- **Monitoring:** New logs are emitted to `/var/log/api_client.log` with rotation every 10MB.

**Next Steps:**
1. Update documentation in `/docs/api_usage.md` to reflect async usage.
2. Run integration tests in the staging environment.
3. Deploy to production in a phased rollout (10% traffic first).