AI-Powered Document Generator

A Visual Studio Code extension that uses AI (Google's Gemini) to automatically generate comprehensive documentation for your code.
This extension can analyze any code files and create well-structured, professional documentation in Markdown format.

✨ Features

• Multi-Language Support
  Generate documentation for any programming language including JavaScript, TypeScript, Python, Java, C#, and more.

• Smart Analysis
  Uses Google's Gemini AI (via official @google/genai library) to understand your code's structure, purpose, and relationships.

• Comprehensive Documentation
  Creates detailed documentation including:

  • Code overview and architecture
  • Dependencies and setup requirements
  • Detailed implementation explanations
  • Usage guides and examples
  • Best practices and recommendations

• Customizable Documentation
  Multiple ways to generate documentation:

  • Predefined templates for common scenarios (general code, tests, APIs, components)
  • Custom prompts for tailored documentation needs
  • Automatic file content inclusion in generated documentation

• Customizable Templates
  Different documentation templates for various types of code:

  • General code documentation
  • Test automation documentation
  • API documentation
  • Component documentation

• Markdown Output
  Clean, well-structured documentation in Markdown format.

• VS Code Integration
  Easy-to-use interface integrated into the VS Code activity bar.

• Folder Tree View
  Visual file selector that mirrors the VS Code Explorer, with expandable/collapsible folders and multi-file selection.

• Customizable File Type Filtering
  Easily adjust which file types are included by editing config/validFiletypes.js in the extension code.

⚙️ Requirements

• Visual Studio Code v1.102.0 or higher
• Node.js (v18+ recommended)
• Google Gemini API key

🚀 Setup

1️⃣ Install the extension from the VS Code Marketplace or .vsix package.

2️⃣ Get your Gemini API key:

• Go to Google AI Studio.
• Copy your Gemini API key.

3️⃣ Set the API key in the extension:

• Method 1 (preferred):

  • Open the extension tab on VS Code.
  • Click Configuration on the top of the widget.
  • Set your Gemini API key.
  • Click Save API Key button.

• Method 2:

  • Open VS Code Settings → search for AI Documentation Generator.
  • Paste your Gemini API key into the geminiApiKey field.

4️⃣ Open the AI Documentation Generator view from the VS Code activity bar.

5️⃣ Select:

• Files you want to document.
• Documentation type (from the dropdown).

6️⃣ Click Generate Documentation and let the extension create beautiful Markdown docs.

🔨 Usage

1. Open your project in VS Code
2. Click the AI Document Generator icon in the Activity Bar
3. Select files you want to document using the folder tree
4. Choose a documentation type:
   • Use predefined templates for common scenarios
   • Select "Custom Prompt" to write your own prompt
   • Custom prompts will automatically include the content of selected files
5. Click "Generate Documentation"

Using Custom Prompts

When selecting "Custom Prompt" from the documentation type dropdown, you can:

1. Enter any prompt you want the AI to follow
2. Your prompt will be automatically combined with the content of selected files
3. Use natural language to describe exactly what kind of documentation you need
4. The extension will maintain proper formatting and structure

Example custom prompt:

Generate a technical specification document that includes:
1. System Overview
2. Architecture Details
3. Implementation Guidelines
4. Security Considerations
5. Performance Requirements

🛠 Technical Notes

• Uses the official @google/genai package.
• Passes the API key correctly via { apiKey } object.
• Does not rely on .env files inside the packaged extension — keys are set via VS Code settings or system env.
• Uses config/validFiletypes.js to define the list of supported file extensions.
• Builds the folder tree dynamically, showing paths relative to the open workspace root (not full system paths).

💬 Questions or Feedback?

Check out the project on GitHub
or reach out to the creator, Douglas D'Avila.

🛡️ Note:
Your API key is stored only in your local VS Code settings and is never transmitted to the extension owner or third parties.