Supabase Types Added

• Added Supabase types for stricter typing and better consistency across the app.

Enhanced Input Forms

• Upgraded the automatically generated input forms of the demo apps to be more beautiful.

Improved App Info Cards

• Redesigned app info cards for a more appealing and professional look.

File Uploads Tracking

• Added file_uploads table to track all file uploads (images, audio, PDFs) across apps.

• Updated useFileUpload hook and relevant API routes.

/lib Folder Reorganized

• Restructured /lib for better organization:

  • /lib/clients for all clients.

  • /lib/ai for AI utilities.

  • /lib/db for database queries.

Next.js Caching for Data Fetching

• Implemented Next.js caching to optimize data fetching across all AI apps.

• Reduced redundant queries by leveraging caching for Supabase requests, improving performance and user experience.

• Works seamlessly with the newly centralized query structure in @/lib/db/queries.
  
  

This update makes data fetching faster and more efficient, enhancing scalability and responsiveness for all demo apps.

Centralized Supabase Queries for All Demo Apps

• Moved all Supabase queries into a centralized directory at @/lib/db/queries for cleaner and more maintainable code.

• Introduced dedicated query files:

  • general.ts: Handles general queries like session state and user credits.

  • chat.ts: Manages all queries for the new chat app.

  • generations.ts: Unified logic for GPT-4.0, DALL-E, SDXL, Llama 3.1, Claude 3.5 Sonnet, and xAI Grok apps.

  • audio.ts: Contains all queries for the audio app.

  • pdf.ts: Handles queries for the PDF app.
    
    

This update simplifies data fetching, reduces redundancy, and improves scalability across all 10 demo apps.

Switched to Slugs for Demo Apps

• Updated demo apps to use slugs instead of IDs in routes and database queries.

• Enables better support for programmatic SEO, allowing cleaner and more descriptive URLs for user-generated content.

• Simplifies navigation and improves readability for both users and search engines.
  
  

This change significantly enhances the SEO potential of demo apps and prepares them for scalable content generation.

Implemented xAI integration through OpenAI SDK

• Implemented through OpenAI SDK.

• Marks the 11th demo app added to @anotherwrapper.

Centralized File Upload Logic

• Introduced a new useFileUpload hook to handle file uploads for images, audio, PDFs, and other file types.

• Consolidated file upload logic into a single, secure implementation used by all API routes.

• Simplified the codebase, making it easier to manage and extend upload functionality across different apps.

Improved API Route Documentation

• Added detailed comments to all API routes, explaining their purpose, logic, and key functionality at the top of each file.

• Improved clarity and readability for developers to better understand the flow and responsibilities of each route.

Improved Webhook API Route Documentation

• Added more detailed comments to the LemonSqueezy and Stripe payment webhook API routes to clarify its functionality.

• Made it easier to understand the purpose of each section and how webhook events are processed.

Centralized Authentication Middleware

• Created a new authMiddleware to centralize authentication logic, reducing code duplication across API routes.

• Applied the middleware to all API routes, ensuring consistent and secure authentication handling.

• Streamlined the codebase for improved maintainability and scalability.
  
  

This update simplifies authentication management while enhancing security across the app.

Stripe Payments Added

• Implemented Stripe payments in the premium repo, fully integrated with the existing credits system.

• Offers users the flexibility to choose between Stripe or Lemonsqueezy as their preferred payment provider.

• Documentation updated to reflect the new feature: Stripe Payments Docs.
  
  

Happy building! 🫡

Optimized PDF Vectorization Process for Chat with PDF demo app

• Implemented batch embedding for improved performance and efficiency during vectorization.

• Enhanced error handling and input validation to ensure smoother processing and better stability.

• Improved text sanitization to handle non-printable characters while preserving common whitespace.

• Added detailed logging for easier debugging and tracking issues during processing.

• Optimized the Supabase insertion process for better database performance.
  
  

These changes should improve the reliability and efficiency of the PDF vectorization workflow.

Console Errors Fix

• Resolved React SVG errors by updating all instances of fill-rule, clip-rule, stroke-width, etc., to fillRule, clipRule, and strokeWidth for proper attribute casing.

• Addressed Punycode deprecation warning in the Supabase package by implementing a workaround based on this.

• These fixes clean up the console logs, ensuring a smoother development experience and fewer distractions during debugging.

Made demo apps programmatic SEO-ready

• Updated the schema and database to store title, subtitle, and description for each generation.

• Adjusted the generations page ([id]/page.tsx) to dynamically fetch and display this metadata.

• Meta tags are now generated based on user-created content for better SEO optimization.

• Removed unnecessary hooks and routes (getAIresponse.ts and fetch/route.ts), streamlining the backend logic.

• Suggested strategy: Use internal backlinking for user-generated content pages to improve search engine crawlability.
  

This update makes programmatic SEO easier and opens up new possibilities for leveraging user-generated content. 🔥

Generation History for Text-Based Generations (LLMs)

• Added a generations history feature to the dashboard for GPT, Llama, Claude, and Vision apps.

• Allows users to view their previous text-based generations directly within the dashboard.

• Improved the UX by centralizing and simplifying access to past content, enhancing productivity and tracking.

Big update to folder structure

• Restructured app organization: Front-end and API routes for demo apps are now neatly organized under a unified (apps) directory for better logic and clarity.

• Unified dashboard: Consolidated all apps (except GPT and Chat) into a single, streamlined dashboard, enhancing navigation and user experience.

• Repo updates: The GitHub repository has been updated to reflect these changes, ensuring alignment with the new structure.
  
  

Try out the new setup on anotherwrapper.com! Documentation updates to follow soon.

UI adjustments for demo apps on the unified dashboard

• Refactored layouts to ensure consistent styling and alignment across all demo apps.

• Updated footer implementation to dynamically use toolConfig for better configurability and cleaner code.

• Improved overall dashboard aesthetics to enhance user experience.

Codebase cleanup: Removed unused layouts

• Deleted outdated and unused layout-old.tsx files across multiple apps, including audio, llama, and DALL-E.

• Streamlined the codebase, reducing unnecessary clutter and improving maintainability.

• Ensured a cleaner structure aligned with the new unified dashboard architecture.

Major update: Unified Dashboard for All AI Apps

• Consolidated 10 separate AI apps into a single, unified dashboard.

• Centralized access to text, image, PDF, audio, and other capabilities, eliminating the need to navigate between multiple UIs.

• Streamlined the user experience for faster and smoother building and testing of AI apps.

• Designed with clean, intuitive UX to enhance productivity and reduce friction.
  
  

This is the biggest upgrade to anotherwrapper.com yet! 🚀

Image generation history added to DALL-E and SDXL apps.

• Introduced a dashboard to display the history of previously generated images for DALL-E and SDXL demo apps.

• Users can now view and manage their past generations, improving usability and tracking.

• Added support for fetching and displaying generation metadata in the UI.

Sanitization added to Chat with PDF app.

• Implemented text sanitization before vectorizing PDFs to prevent issues with non-printable characters or extra whitespace.

• Ensured cleaner and more robust embeddings by handling invalid characters during processing.

Upgraded all GPT-3.5 demo apps to new GPT-4o-mini.

• Switched from GPT-3.5-turbo-1106 to the newly released GPT-4o-mini for all relevant demo apps.

• Ensured immediate adoption of OpenAI's latest model for enhanced performance and accuracy.

• Updated app descriptions to reflect the transition to GPT-4o-mini. 🚀

Supabase local DB schema support added.

• Added SQL schema to the repository, allowing users to push their database using the Supabase CLI without manual copy-pasting.

• Updated to the latest Next.js v14.2.5 for improved performance and compatibility.

Added safety check to redirect.

• Updated generateAIResponse to include a fallback mechanism for toolConfig.toolPath.

• Ensured safer redirection logic by dynamically verifying the existence of toolPath before constructing the redirect path.

• Made systemMessage optional in generateVisionResponse for greater flexibility in usage.

New AI App: Text-to-Speech in 26 Languages with 1,000+ Voices! 🔥

• Just added the 10th AI demo app, introducing lifelike voice generation.

• Supports 26 languages and 1,000+ unique voices, powered by ElevenLabs integration.

• Adds AI voice capabilities on top of existing AI text, image, audio, and vision features already available.

Google OAuth flow refactored.

• Moved the redirectTo logic into a centralized function using getRedirectUrl for dynamic URLs based on environment.

• Replaced direct Supabase client calls in the GoogleSignInButton with an API route (`/api/auth/google`) for better separation of concerns.

• Enhanced maintainability and debugging by centralizing the authentication logic in the API route.

Emails converted to lowercase on purchase.

• Ensured all email comparisons use toLowerCase() for consistency.

• Prevented potential mismatches caused by email casing in profile queries and updates.

Chat Model Configuration Update

• Replaced hardcoded model (`gpt-3.5-turbo-1106`) with a dynamic model reference from toolConfig.aiModel.

• Ensures flexibility and consistency by using the centralized configuration for the chat model.

Improved Contact Creation for Loops

• Updated logic to allow adding contacts to Loops without requiring a name.

• Ensures contacts are still created using just an email while dynamically adding a name if available.

Claude 3.5 Model Update

• Upgraded AI model from claude-3-20240229 to claude-3-5-sonnet-20240620 for enhanced performance and functionality.

Product growth update:

• Got first couple of sales

• Added new demo apps

• Posted on different directories

• Closed sponsorship deal with AIDirectories

Dynamic OG Banners

Implemented dynamic Open Graph banners using Next.js OG for better link previews and enhanced sharing experience. 🚀🎨