Welcome to RevivAI! This guide will walk you through all the main features and help you get started with AI-powered code documentation. RevivAI makes it easy to understand, document, and interact with your codebase using advanced language models.
Note
This guide assumes you have RevivAI running locally or have access to a hosted instance. Check the README.md for installation instructions.
- Creating a New Project
- Setting Up LLM Providers
- Uploading Your Code
- Understanding the Interface
- Chatting with Your Codebase
- Generating Documentation
- Settings
Projects in RevivAI help you organize your work. Each project contains one codebase and all associated chat history and documentation.
-
Start from the Home Page: When you first open RevivAI, you'll see the projects page with a list of existing projects (if any).
-
Click "New Project": Click "New Project" button to open the creation dialog.
-
Enter Project Name: Give your project a descriptive name like "My Graduation Project" or "Company API Documentation".
-
Setup LLM Provider: If you haven't set up an AI provider yet, you'll be redirected to the setup page after creating the project.
Tip
Use descriptive project names to easily identify different codebases later. Consider including the project type or version in the name.
- Browse Projects: All your projects are listed on the home page, sorted by creation date
- Open Project: Click on any project name to open it
- Delete Project: Hover over a project and click the trash icon to delete it
- Project Info: Each project shows when it was created
RevivAI supports multiple AI providers. You need to configure at least one to use the app's features.
RevivAI supports these LLM providers:
- OpenAI (GPT-4, GPT-3.5, etc.)
- Anthropic (Claude models)
- Google (Gemini models)
- OpenRouter (Access to multiple models)
- Ollama (Local/self-hosted models)
- Custom API (Any OpenAI-compatible API)
-
Access Setup: Go to setup from the home page when creating a project, or click "Go to Setup" in project settings.
-
Choose Provider: Select your preferred AI provider from the tabs at the top.
-
Configure Settings:
- API URL: The endpoint for your chosen provider (pre-filled for popular services)
- Model: Specify the model name (e.g., "gpt-4", "claude-3-sonnet")
- API Key: Your API key for the service
-
Test Connection: Click "Test" to verify your settings work correctly.
-
Save: Click "Save & Continue" to apply your configuration.
Each provider has different steps to get an API key:
- OpenAI: Visit the OpenAI Developer Portal
- Anthropic: Go to the Anthropic Console
- Google: Access Google AI Studio
- OpenRouter: Check OpenRouter Settings
Important
Most providers require a paid account, though some offer free tiers with limitations.
All API calls are made directly from your browser to the AI provider. RevivAI never stores or sees your API keys or data.
Before you can chat with your code or generate documentation, you need to upload your codebase to RevivAI.
RevivAI supports two ways to upload code:
- Select Folder: Click the upload button and select a folder containing your project
- File Filtering: Use include/ignore patterns to control which files are uploaded
- Size Limit: Individual files larger than 20MB are automatically skipped
- Public Repositories: Enter any public GitHub repository URL
- Automatic Download: RevivAI will fetch and process the repository contents
Control which files are included using glob patterns:
Specify which files to include (comma-separated):
src/**/*.js, *.md, package.json
Specify which files to exclude (comma-separated):
node_modules/**, *.log, dist/**, .git/**
- Open Upload Dialog: Click the folder icon in the navigation bar
- Choose Method: Select either "Local Files" or "GitHub Repository"
- Configure Filters: Set include/ignore patterns if needed
- Enable Compression: Check this option to reduce the codebase size. This may slightly decrease documentation quality, but it saves tokens and allows you to process larger codebases.
- Upload: Click "Upload" and wait for processing
After upload, you'll see:
- Total Files: Number of files processed
- Token Count: How many AI tokens your codebase uses
- Largest Files: Which files use the most tokens
- Compression Status: Whether compression was applied
Warning
Large codebases can consume many AI tokens and increase costs. Use compression and filtering to optimize your upload.
- View Current: See details about your uploaded codebase in the upload dialog
- Upload New: Replace the current codebase with a new one
- File Analysis: Review which files use the most tokens
Once you've created a project and uploaded your code, you'll see RevivAI's main interface with a navigation bar at the top containing several important icons and controls.
The navbar provides quick access to all major features:
- Back Button: Returns you to the projects homepage
- Chat/Writer Tabs: Switch between chatting with your code and generating documentation
- Theme Toggle: Switch between light and dark modes for comfortable viewing
- Upload Button: Access the upload dialog to manage your codebase
- Settings Button: Open settings to configure AI providers, language, and generation options
The chat feature lets you ask questions about your code and get AI-powered explanations.
- Open Chat Tab: Click on "Chat" in the navigation bar
- Welcome Message: You'll see a welcome message with suggestions
- Type Questions: Use the text input at the bottom to ask questions
The AI can help with many code-related questions:
- "What does this function do?"
- "How does the authentication system work?"
- "Explain the main components of this app"
- "What's the overall architecture of this project?"
- "How are the files organized?"
- "Show me the data flow through the application"
- "How is user registration handled?"
- "What database models are used?"
- "Explain the API endpoints"
- "Create a diagram showing the class relationships"
- "Generate a flowchart of the user login process"
- "Show the component hierarchy"
- Copy Responses: Click the copy icon to copy AI responses
- Delete Messages: Remove individual messages using the trash icon
- Clear Chat: Use the "Clear Chat" badge to start fresh
- Code Highlighting: Code snippets are automatically highlighted
- Mermaid Diagrams: The AI can generate interactive diagrams
- Markdown Formatting: Responses support full Markdown formatting
- Codebase Context: The AI has access to your entire uploaded codebase
- Previous Messages: Conversations maintain context across messages
- Token Usage: See how many tokens you're using in the info bar
- Be Specific: Ask clear, specific questions for better answers
- Reference Files: Mention specific files or functions when relevant
- Ask for Examples: Request code examples to illustrate concepts
- Request Diagrams: Ask for visual representations of complex concepts
RevivAI can automatically generate comprehensive documentation for your codebase using AI.
Tip
Start with a template to get a structured foundation, then customize the chapters to match your project's specific needs.
- Open Writer Tab: Click on "Writer" in the navigation bar
- Choose Template: Start with a pre-built template or create custom chapters
- Generate Content: Use AI to automatically write documentation
RevivAI provides several documentation templates:
- Perfect for open-source projects
- Includes project introduction, features, setup instructions
- Covers usage examples and contribution guidelines
- Automatically analyzes your codebase structure
- Generates a custom documentation outline
- Adapts to your specific project type and architecture
Note
The AI-Assisted template uses your actual code structure to create relevant chapters. Results may vary depending on your codebase complexity.
- Add New Chapter: Click "New Chapter" to create custom sections
- Chapter Titles: Use descriptive titles that will become filenames
- Chapter Outlines: Write detailed outlines in Markdown format
- Title Field: Set the chapter name and filename
- Outline Editor: Write chapter content using Markdown
- Live Preview: See how your content will look
- Chapter Navigation: Click chapter names to switch between them
- Drag & Drop: Reorder chapters by dragging them in the sidebar
- Delete Chapters: Remove unwanted chapters using the delete button
Tip
Write detailed outlines for better AI-generated content. Include specific topics, code examples you want covered, and any diagrams that would be helpful.
- Prepare Chapters: Create chapter outlines describing what you want
- Click Generate: Use the "Generate" button to start AI writing
- Monitor Progress: Watch as chapters are generated in real-time
- Parallel Processing: Multiple chapters can be generated simultaneously
- Progress Tracking: See which chapters are being worked on
- Error Handling: Failed generations can be retried
- Content Quality: AI writes beginner-friendly, comprehensive content
Warning
Generation can consume significant API tokens, especially for large projects. Monitor your usage and consider enabling parallelization limits if you're concerned about costs.
- Download ZIP: Get all documentation as a compressed archive
- Preview Files: View generated content before downloading
- Multiple Formats: Files are generated as Markdown (.md) files
- Generation History: See all previous documentation exports
- Selective Download: Choose which exports to download
- File Preview: View individual files before downloading
- Language Support: Generate documentation in multiple languages
- Code Examples: AI includes relevant code snippets from your codebase
- Diagrams: Automatic generation of Mermaid diagrams and flowcharts
- Beginner-Friendly: Content is written for newcomers to understand
- Technical Accuracy: AI references actual code from your project
- Cross-References: Chapters link to related sections automatically
Customize RevivAI's behavior through the settings panel to optimize performance and output quality.
Note
Settings are saved locally in your browser and persist across sessions. Changes take effect immediately.
Click the settings icon (sliders) in the navigation bar to open the settings dialog.
- Purpose: Controls the language used for AI responses and documentation
- Default: English
- Options: Multiple languages supported including Spanish, French, German, and more
- Impact: Affects both chat responses and generated documentation
- Purpose: Controls AI creativity and randomness
- Range: 0.0 to 2.0
- Default: 0.7
- Lower Values (0.0-0.5): More deterministic, focused responses
- Higher Values (1.0-2.0): More creative, varied responses
- Purpose: Controls how many chapters are generated simultaneously
- Options:
- Disabled (1): Generate one chapter at a time
- 2, 4, 8: Generate multiple chapters in parallel
- Unlimited: Maximum parallel processing
- Benefits: Faster generation with higher values
- Caution: May trigger API rate limits with very high values
Warning
Setting parallelization too high can trigger rate limits from your AI provider, causing generation failures. Start with lower values and increase gradually.
- Access: Click "Go to Setup" in settings
- Purpose: Switch between different AI providers or models
- Redirection: Takes you back to the setup page
- Preserves: Your current project data remains intact
- Different Models: Each provider offers different capabilities
- Cost Optimization: Some providers may be more cost-effective
- Performance: Different models excel at different tasks
Tip
Experiment with different providers to find the best balance of cost, speed, and quality for your specific needs. Some models are better at code analysis, others at creative writing.
- Location: Available in project settings
- Warning: This action permanently deletes all project data
- Includes: Removes codebase, chat history, and generated documentation
- Confirmation: Requires explicit confirmation before deletion
- Monitor Usage: Check token counts in the info bar during chat
- File Filtering: Use include/ignore patterns to reduce codebase size
- Compression: Enable compression when uploading large codebases
- Parallelization: Increase for faster generation (if your API allows)
- Temperature: Lower values for more consistent documentation
- Chapter Size: Break large topics into smaller, focused chapters
- Rate Limits: Reduce parallelization if hitting API limits
- Quality Issues: Adjust temperature for better content quality
- Language Problems: Ensure language setting matches your preference
- Performance: Enable compression and optimize file filtering
Note
RevivAI is designed to be intuitive and user-friendly. Most features work automatically once you've uploaded your codebase and configured an AI provider.