Local LLM Chat for Visual Studio

Features

🤖 Local AI Chat

• Privacy-First: All data stays on your machine—no cloud APIs required
• OpenAI-Compatible: Works with Ollama, LM Studio, OpenAI, and any OpenAI-compatible endpoint
• Persistent Conversations: Chat history is maintained throughout your Visual Studio session
• Integrated Tool Window: Chat interface seamlessly integrated into Visual Studio's UI

📁 Solution Integration

• Read Files: Load any solution file into the conversation context
• List Directories: Browse your project structure directly from chat
• Search Files: Find files using wildcard patterns (e.g., *.cs, *.json)
• Solution Info: Get metadata about your solution (Git status, Node.js projects, etc.)
• Smart Context: Send active file to the AI with a single command

✨ File Creation & Management

• AI-Powered File Generation: Let the AI create complete files based on your requirements
• Special Syntax: AI responds with ```file path="relative/path.ext" blocks for file suggestions
• Safe Operations: Confirmation prompts before creating/overwriting files
• Path Validation: Security checks prevent file access outside your solution directory

⚡ Chat Commands

Built-in slash commands for quick actions:

• /read <file-path> - Read a file and add to conversation
• /list [directory] - List files in a directory
• /search <pattern> - Search for files with wildcard patterns
• /workspace - Show solution information
• /help - Display all available commands

Installation

Requirements

• Visual Studio 2022 (version 17.0 or higher) - Community, Professional, or Enterprise
• .NET Framework 4.7.2 or higher
• Local LLM Server: Ollama, LM Studio, or compatible OpenAI API endpoint

From VSIX File

1. Download the .vsix file from the releases page
2. Close Visual Studio if running
3. Double-click the .vsix file to install
4. Or use the command line: VSIXInstaller.exe /quiet LocalLLMChatVS.vsix
5. Restart Visual Studio

From Source

git clone https://github.com/markusbegerow/local-llm-chat-vs.git
cd local-llm-chat-vs

Open LocalLLMChatVS.sln in Visual Studio 2022, then press F5 to build and launch in the experimental instance.

Getting Started

1. Set Up Your Local LLM

Option A: Ollama (Recommended)

# Install Ollama from https://ollama.ai
ollama pull llama3.2
ollama serve

Option B: LM Studio

1. Download from lmstudio.ai
2. Load a model
3. Start the local server (default: http://localhost:1234/v1/chat/completions)

Option C: OpenAI API

• Use directly with your OpenAI API key
• Endpoint: https://api.openai.com/v1/chat/completions

2. Configure the Extension

1. Open Visual Studio 2022
2. Go to Tools → Options → Local LLM Chat → General
3. Configure the following settings:
   • API URL: Full endpoint URL (e.g., http://localhost:11434/v1/chat/completions)
   • API Token: Your API token or "ollama" for local Ollama
   • Model Name: The model to use (e.g., llama3.2, gpt-4)

Example Ollama Configuration:

API URL: http://localhost:11434/v1/chat/completions
API Token: ollama
Model Name: llama3.2

Example OpenAI Configuration:

API URL: https://api.openai.com/v1/chat/completions
API Token: sk-your-api-key-here
Model Name: gpt-4

3. Start Chatting

1. Open a solution in Visual Studio
2. Go to View → Other Windows → Local LLM Chat
3. Or use the command: Tools → Local LLM Chat → Open Chat
4. Start asking questions!

Usage Examples

Analyze Code

/read Controllers/HomeController.cs
Can you explain how this controller works and suggest improvements?

Explore Project Structure

/workspace
/list Controllers
What is the architecture of this project?

Generate Files

Create a C# service class for user authentication with JWT tokens, including proper error handling and XML documentation.

The AI will suggest a file with path and content. Click "Yes" in the confirmation dialog to create it!

Search and Refactor

/search *.cs
Find all C# files, then help me refactor the error handling patterns across the project.

Quick File Analysis

1. Open any file in the editor
2. Go to Tools → Local LLM Chat → Send Active File to Chat
3. Ask questions about the code

Send Specific File

1. Go to Tools → Local LLM Chat → Send File to Chat
2. Select the file from the dialog
3. The file content will be added to the conversation context

Commands

Tools Menu Commands

Access these commands from Tools → Local LLM Chat menu:

You can also access the chat window from View → Other Windows → Local LLM Chat

In-Chat Slash Commands

Note: Press Ctrl+Enter in the chat input to send your message.

Configuration

Access settings through Tools → Options → Local LLM Chat → General

API Configuration

Model Parameters

Conversation

Network

Security

Configuration Examples

Ollama (Local)

API URL: http://localhost:11434/v1/chat/completions
API Token: ollama
Model Name: llama3.2

LM Studio (Local)

API URL: http://localhost:1234/v1/chat/completions
API Token: lmstudio
Model Name: your-model-name

OpenAI (Cloud)

API URL: https://api.openai.com/v1/chat/completions
API Token: sk-your-actual-api-key-here
Model Name: gpt-4

Security Features

• ✅ Path Traversal Protection: Prevents access outside solution directory
• ✅ File Size Limits: Configurable maximum file sizes (default: 1MB)
• ✅ Path Validation: All file paths validated before operations
• ✅ URL Validation: API endpoints validated for security
• ✅ Confirmation Prompts: Review before creating/overwriting files
• ✅ Local-Only Default: No external API calls unless you configure them
• ✅ Binary/Build Exclusion: Automatically excludes bin/, obj/, node_modules/, .git/

Requirements

• Visual Studio 2022: Version 17.0 or higher (Community, Professional, or Enterprise)
• .NET Framework: Version 4.7.2 or higher
• Local LLM: Ollama, LM Studio, or compatible OpenAI API server

Development

Building from Source

# Clone repository
git clone https://github.com/markusbegerow/local-llm-chat-vs.git
cd local-llm-chat-vs

1. Open LocalLLMChatVS.sln in Visual Studio 2022
2. Restore NuGet packages (right-click solution → Restore NuGet Packages)
3. Build the solution (Ctrl+Shift+B)
4. Press F5 to launch in experimental instance

Packaging as VSIX

1. Build the project in Release mode
2. The VSIX file will be generated in bin\Release\LocalLLMChatVS.vsix
3. Install it by double-clicking the VSIX file

Project Structure

LocalLLMChatVS/
├── Commands/
│   ├── OpenChatCommand.cs           # Open chat window command
│   ├── ClearConversationCommand.cs  # Clear chat command
│   ├── SendFileToChatCommand.cs     # Send file command
│   └── SendActiveFileToChatCommand.cs # Send active file command
├── Services/
│   ├── LLMService.cs                # LLM API integration
│   └── WorkspaceService.cs          # Solution/file operations
├── ToolWindows/
│   ├── ChatWindow.cs                # Chat tool window
│   └── ChatWindowControl.xaml.cs    # Chat UI control (WPF)
├── Models/
│   └── ChatMessage.cs               # Data models
├── Utilities/
│   ├── PathValidator.cs             # Path security validation
│   └── SecurityValidator.cs         # Security utilities
├── Options/
│   └── GeneralOptions.cs            # Settings page
├── LocalLLMChatPackage.cs           # Main package class
├── source.extension.vsixmanifest    # Extension manifest
└── VSCommandTable.vsct              # Command definitions

Troubleshooting

Connection Issues

• Verify LLM server is running:
  • Ollama: curl http://localhost:11434/v1/models or ollama list
  • LM Studio: Check the server tab shows "Server Running"
• Check API URL in settings:
  • Go to Tools → Options → Local LLM Chat → General
  • Verify the API URL matches your server
  • Ensure the endpoint includes /v1/chat/completions
• Test with curl:

  curl http://localhost:11434/v1/models
  

Chat Not Responding

• Check error messages in the chat window or Visual Studio Output window
• Increase timeout: Go to Tools → Options → Local LLM Chat → General and increase "Request Timeout"
• Verify model is loaded in your LLM server
• Check API token is correct (use "ollama" or any dummy value for local servers)

File Operations Failing

• Ensure a solution is open in Visual Studio
• Check file paths are relative to the solution directory (no absolute paths)
• Verify file size limits in Tools → Options → Local LLM Chat → Security
• Check permissions on the solution directory

Extension Not Loading

• Check Visual Studio version: Must be 2022 (17.0+)
• Verify .NET Framework 4.7.2 or higher is installed
• Check Extensions → Manage Extensions to see if extension is enabled
• Try resetting Visual Studio settings: Tools → Import and Export Settings

Recommended Models

Coding Tasks (C# / .NET)

• Qwen 2.5 Coder (7B/14B): Excellent for C# and .NET development
• CodeLlama (13B/34B): Specialized for code generation and understanding
• DeepSeek Coder (6.7B/33B): Strong at complex algorithms and refactoring
• Llama 3.2 (3B/8B): Good balance of speed and capability

General Tasks

• Llama 3.2: Best all-around performance for most tasks
• Mistral 7B: Fast and efficient for quick questions
• Phi-3: Compact but capable for lighter workloads

Cloud/Enterprise

• GPT-4 / GPT-4 Turbo: Best quality, requires OpenAI API key
• GPT-3.5 Turbo: Faster and cheaper alternative

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Development Guidelines

• Follow C# coding conventions
• Test in Visual Studio 2022 experimental instance
• Update documentation for new features
• Ensure security validators are used for file operations

License

This project is licensed under the GPL License - see the LICENSE.txt file for details.

Acknowledgments

• Thanks to the Ollama team for making local LLMs accessible
• LM Studio for providing an excellent local inference platform
• The Visual Studio extensibility team for comprehensive SDK and documentation
• OpenAI for the standardized API format

🙋‍♂️ Get Involved

If you encounter any issues or have questions:

• 🐛 Report bugs
• 💡 Request features
• ⭐ Star the repo if you find it useful!

☕ Support the Project

If you like this project, support further development with a repost or coffee:

📬 Contact

• 🧑‍💻 Markus Begerow
• 💾 GitHub
• ✉️ Twitter

Privacy Notice: This extension operates entirely locally by default. No data is sent to external servers unless you explicitly configure it to use a remote API endpoint (like OpenAI).