Skip to content

Latest commit

 

History

History
95 lines (73 loc) · 2.58 KB

CHANGES.md

File metadata and controls

95 lines (73 loc) · 2.58 KB

Changes to kubectl-mcp-tool Configuration

Key Improvements

  1. Simplified MCP Server Implementation

    • Created a minimal wrapper (kubectl_mcp_tool.minimal_wrapper.py) for better compatibility
    • Removed complex parameter schemas that were causing compatibility issues
    • Used direct tool registration with simple names

  2. Improved Cursor Integration

    • Updated Cursor configuration to use the minimal wrapper
    • Added explicit environment variables for PATH and KUBECONFIG
    • Provided better error handling and logging

  3. Enhanced Claude and WindSurf Support

    • Updated configuration for Claude Desktop
    • Updated configuration for WindSurf
    • Standardized configuration format across all AI assistants

  4. Streamlined Installation Process

    • Improved installation script to set up all configurations automatically
    • Added automatic testing of kubectl and kubeconfig
    • Created better error handling during installation

  5. Comprehensive Documentation Updates

    • Updated README.md with working configuration examples
    • Updated integration guides for all supported AI assistants
    • Created a new QUICKSTART.md guide for new users
    • Enhanced troubleshooting section with more specific solutions

Configuration Changes

Previous Configuration (Not Working)

{
  "mcpServers": {
    "kubernetes": {
      "command": "python",
      "args": ["-m", "kubectl_mcp_tool.cli.cli"]
    }
  }
}

New Configuration (Working)

{
  "mcpServers": {
    "kubernetes": {
      "command": "python",
      "args": ["-m", "kubectl_mcp_tool.minimal_wrapper"],
      "env": {
        "KUBECONFIG": "/path/to/your/.kube/config",
        "PATH": "/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/homebrew/bin"
      }
    }
  }
}

Implementation Details

The key implementation changes include:

  1. Using a simpler tool registration approach:
@server.tool("process_natural_language")
async def process_natural_language(query: str):
    # Implementation...

  1. Avoiding complex parameter schemas that aren't supported in some MCP SDK versions

  2. Setting explicit environment variables to ensure kubectl can be found

  3. Better error handling throughout the codebase

  4. Comprehensive debugging and logging

Testing Your Installation

After applying these changes, you can verify your installation with:

# Test command line
kubectl-mcp --help

# Test MCP server directly
python -m kubectl_mcp_tool.minimal_wrapper

# Run automated installation
bash install.sh

Then try using your AI assistant to interact with your Kubernetes cluster.