charris-msft/chat-mode-creator-mcp

🎯 Chat Mode Creator MCP Server

An advanced MCP (Model Context Protocol) server that creates custom VS Code chat modes AND custom prompts for specialized development workflows.

✨ Key Features

🎯 Integrated Workflow: Chat modes come with related custom prompts
📁 VS Code Compatible: Follows official VS Code chat mode and prompt specifications
🚀 Ready-to-Use: Pre-built templates for common development scenarios
⚡ Intelligent Matching: Analyzes queries to suggest appropriate modes

🛠️ Available Chat Modes & Custom Prompts

🔄 Lambda to Azure Functions Migration

Chat Mode: Comprehensive migration guidance from AWS Lambda to Azure Functions

• Runtime and service mapping assistance
• Code conversion patterns and best practices
• Infrastructure migration support

Custom Prompts:

• /evaluate - Assess migration readiness and complexity
• /validate - Verify migrated functions work correctly

🏗️ Azure Bicep Development

Chat Mode: Expert assistance for Azure Bicep infrastructure as code

• Azure Well-Architected Framework integration
• Advanced Bicep patterns and techniques
• Security, monitoring, and deployment best practices

Custom Prompts:

• /review - Code review and best practices validation
• /deploy - Safe deployment procedures and checks

📋 Spec-Driven Development

Chat Mode: Transform specifications into code through systematic process

• Treats specifications as the primary development artifact
• Follows systematic 5-phase SDD workflow
• Includes consistency validation and checklist integration

Custom Prompts:

• /validate_spec - Validate specifications for completeness
• /generate_tests - Generate comprehensive test suites

🚀 Quick Installation

1. Install Dependencies

# Create and activate virtual environment (.venv)
python -m venv .venv
.\.venv\Scripts\Activate.ps1  # Windows PowerShell
# .venv/bin/activate          # Linux/Mac

# Install required packages
pip install -r requirements.txt

2. Test Server

# Test that the server runs correctly
python mcp_chatmode_server.py

3. Configure VS Code

Add to your VS Code User Settings (settings.json):

{
  "mcp": {
    "servers": {
      "chat-mode-creator": {
        "command": "C:\\path\\to\\your\\.venv\\Scripts\\python.exe",
        "args": ["C:\\path\\to\\your\\mcp_chatmode_server.py"],
        "cwd": "C:\\path\\to\\your\\workspace",
        "env": {
          "PYTHONPATH": "C:\\path\\to\\your\\workspace",
          "VIRTUAL_ENV": "C:\\path\\to\\your\\.venv"
        }
      }
    }
  },
  "chat.mcp.discovery.enabled": true,
  "chat.promptFiles": true
}

🛠️ Available Tools

create_chat_mode ⭐ Main Tool

Creates chat modes with integrated custom prompts

• Parameters:
  • mode_type (required): "spec_driven_development", "lambda_to_functions_migration", or "azure_bicep_development"
  • workspace_path (optional): Workspace path (defaults to current directory)

suggest_chat_mode 🎯 Smart Assistant

Analyzes your query and suggests the most appropriate chat mode

• Parameters:
  • user_query (required): Description of what you want to accomplish
  • workspace_path (optional): Workspace path (defaults to current directory)

create_custom_prompt 🎨 Advanced

Create individual custom prompts for specific tasks

• Parameters:
  • mode_type (required): The chat mode this prompt belongs to
  • prompt_name (required): Name for the custom prompt
  • custom_description (optional): Custom description for the prompt

list_mode_prompts 📋 Reference

List all available custom prompts for a specific chat mode

• Parameters:
  • mode_type (required): The chat mode to list prompts for

📁 File Structure

Files are created following VS Code standards:

.github/
├── chatmodes/
│   ├── azure-bicep-development.chatmode.md
│   ├── lambda-to-functions-migration.chatmode.md
│   └── spec-driven-development.chatmode.md
└── prompts/
    ├── review.prompt.md      # Azure Bicep prompts
    ├── deploy.prompt.md
    ├── evaluate.prompt.md    # Lambda migration prompts  
    ├── validate.prompt.md
    ├── validate_spec.prompt.md   # Spec-driven prompts
    └── generate_tests.prompt.md

💡 Usage Examples

Natural Language Interaction

Just ask Copilot naturally:

• "I want to migrate my AWS Lambda to Azure Functions"
• "Help me with Azure Bicep infrastructure templates"
• "I need assistance with specification-driven development"

Copilot will automatically use the MCP tools to suggest and create appropriate chat modes.

Direct Tool Usage

# Create Azure Bicep mode with prompts
create_chat_mode(
    mode_type="azure_bicep_development",
    workspace_path="/path/to/your/project"
)

# Get intelligent suggestions
suggest_chat_mode(
    user_query="I need help migrating serverless functions",
    workspace_path="/path/to/your/project"
)

# Create a custom prompt
create_custom_prompt(
    mode_type="azure_bicep_development",
    prompt_name="security_scan",
    custom_description="Scan templates for security issues"
)

🎯 Using in VS Code

1. Chat Modes

• Open VS Code Chat view
• Select your custom mode from the dropdown
• Get specialized assistance for that workflow

2. Custom Prompts

• In chat, type /prompt_name (e.g., /review, /evaluate)
• Prompts automatically include mode-specific context
• Combine with file references for targeted help

3. Integration Example

Chat Mode: Azure Bicep Development
Prompt: /review
Context: main.bicep file
Result: Comprehensive security and best practices review

🔧 Troubleshooting

MCP Server Won't Connect in VS Code

1. Check VS Code MCP settings:

   • Ensure "chat.mcp.discovery.enabled": true is set
   • Verify the server paths are correct with capital drive letters
   • Check that the virtual environment path exists

2. Restart VS Code after making settings changes

3. Check the MCP server logs in VS Code:

   • Open Command Palette (Ctrl+Shift+P)
   • Run "Developer: Show Logs"
   • Look for MCP server connection errors

Server Won't Start

1. Ensure virtual environment is activated:

   .\.venv\Scripts\Activate.ps1  # Windows
   source .venv/bin/activate     # Linux/Mac
   

2. Check dependencies:

   pip install -r requirements.txt
   

3. Test manually:

   python mcp_chatmode_server.py
   

Copilot Not Using MCP Tools

1. Make sure MCP discovery is enabled: "chat.mcp.discovery.enabled": true
2. Try explicit requests: "Use the suggest_chat_mode tool to help me..."
3. Check server status: Look for "chat-mode-creator" in the MCP server list
4. Restart VS Code if the server was recently added

Environment Variables Issues

If you see import errors, ensure these environment variables are set in your MCP configuration:

"env": {
  "PYTHONPATH": "/path/to/your/workspace",
  "VIRTUAL_ENV": "/path/to/your/.venv"
}

🎨 Advanced Configuration

Alternative Launcher Scripts

You can also use PowerShell scripts for easier management:

run-mcp-server.ps1:

cd "C:\path\to\your\workspace"
.\.venv\Scripts\Activate.ps1
python mcp_chatmode_server.py

VS Code Configuration with PowerShell:

{
  "mcp": {
    "servers": {
      "chat-mode-creator": {
        "command": "powershell",
        "args": [
          "-ExecutionPolicy", "Bypass",
          "-File", "C:\\path\\to\\run-mcp-server.ps1"
        ]
      }
    }
  }
}

Claude Desktop Integration

For Claude Desktop users, add to your configuration:

{
  "mcpServers": {
    "chat-mode-creator": {
      "command": "/path/to/your/.venv/bin/python",
      "args": ["/path/to/your/mcp_chatmode_server.py"],
      "cwd": "/path/to/your/workspace"
    }
  }
}

🎯 Custom Prompt Examples

Lambda Migration Evaluation (/evaluate)

Analyzes AWS Lambda functions for migration readiness:

• Runtime compatibility assessment
• Infrastructure dependency mapping
• Performance characteristic evaluation
• Migration complexity scoring

Bicep Template Review (/review)

Comprehensive Bicep template analysis:

• Security best practices validation
• Azure Well-Architected Framework compliance
• Cost optimization recommendations
• Code quality assessment

Specification Validation (/validate_spec)

Ensures specifications are ready for code generation:

• Completeness checking
• Consistency analysis
• Implementation readiness validation
• Gap identification and resolution

🤝 Contributing

Extend this MCP server with additional chat modes:

1. Add new templates to chat_modes/ directory
2. Create new prompt files for the mode
3. Update AVAILABLE_PROMPT_TEMPLATES in server code
4. Test with the provided test scripts
5. Submit PR with examples and documentation

Template Structure

---
description: Brief description of the chat mode
tools: ['relevant', 'tools', 'list']
---

# Chat Mode Name

## Purpose
What this mode helps with...

## Usage Guidelines  
How to use this mode effectively...

## Related Prompts
- `/prompt1` - Description
- `/prompt2` - Description

📊 Project Structure

chat-mode-creator/
├── mcp_chatmode_server.py          # Main MCP server
├── chat_modes/                     # Source templates
│   ├── azure-bicep-development.chatmode.md
│   ├── lambda-to-functions-migration.chatmode.md
│   ├── spec-driven-development.chatmode.md
│   └── */                          # Mode-specific prompts
├── requirements.txt                # Python dependencies
├── example_usage.py                # Usage examples
├── test_server.py                  # Test scripts
└── .venv/                          # Virtual environment

📄 License

MIT License - See LICENSE file for details

🎉 Ready to supercharge your VS Code chat experience with specialized modes and custom prompts!

This MCP server intelligently creates organized chat modes and prompts that follow VS Code's official specifications, making your development workflow more efficient and context-aware.