JSPM

  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 5
  • Score
    100M100P100Q34936F
  • License MIT

A single-file HTML application that allows you to analyze entire folders of files using Large Language Models (LLM). Upload a folder, ask questions, and get AI-powered insights across all your files. Part of a 3-stage loading architecture for maximum reliability.

Package Exports

  • @kosmas10/folder-llm-analyzer
  • @kosmas10/folder-llm-analyzer/folder-llm-analyzer.html

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@kosmas10/folder-llm-analyzer) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

Folder LLM Analyzer

Version: 0.0.4
Package: @kosmas10/folder-llm-analyzer

A single-file HTML application that allows you to analyze entire folders of files using Large Language Models (LLM). Upload a folder, ask questions, and get AI-powered insights across all your files.

🎯 Key Features

  • 📁 Multi-format Support: Analyzes text files (.txt, .md, .json, .csv, .js, .py, etc.) and PDF files
  • ⚡ Batch Processing: Process multiple files simultaneously with progress tracking
  • 📊 Export Results: Export analysis results to Excel (.xlsx) or CSV format
  • 🔄 Real-time Progress: See which file is currently being processed
  • 🛡️ Error Handling: Graceful error handling with detailed error messages
  • 🔔 Version Updates: Automatic version checking and update notifications
  • 💻 No Installation Required: Single HTML file that runs in any modern browser

🏗️ Architecture: 3-Stage Loading System

This application uses an innovative 3-stage loading architecture designed for maximum reliability and flexibility:

Stage 1: Bootstrap (User Installs This)

  • File: folder-llm-analyzer-bootstrap.html
  • Purpose: Ultra-minimal loader that NEVER changes
  • Size: ~2KB
  • What it does: Loads Stage 2 from npm/CDN

Stage 2: Loader (Auto-Updates)

  • Package: @kosmas10/html-app-loader
  • Purpose: Sophisticated HTML app loader
  • What it does: Handles complex loading logic, can be updated via npm

Stage 3: Application (Auto-Updates)

  • Package: @kosmas10/folder-llm-analyzer (this package)
  • Purpose: The actual Folder LLM Analyzer application
  • What it does: Provides the full folder analysis experience

Why This Architecture?

  1. 🛡️ Future-Proof: Stage 1 never changes, so users never need to update their local file
  2. 🔄 Auto-Updates: Stages 2 & 3 update automatically via npm/CDN
  3. 🐛 Bug Fixes: Loader bugs can be fixed without user intervention
  4. 📦 Flexible: Can adapt to new requirements without breaking existing installations

🚀 Quick Start

  1. Download the bootstrap file (one time only):

    curl -O https://cdn.jsdelivr.net/npm/@kosmas10/folder-llm-analyzer@latest/folder-llm-analyzer-bootstrap.html

  2. Open it in your browser:

    • Double-click the file, or
    • Open it from your browser's File menu

  3. That's it! The application will load automatically and stay up-to-date.

Simply navigate to:

https://cdn.jsdelivr.net/npm/@kosmas10/folder-llm-analyzer@latest/folder-llm-analyzer-bootstrap.html

Option C: Load Specific Version

Use URL parameters to control versions:

https://cdn.jsdelivr.net/npm/@kosmas10/folder-llm-analyzer@latest/folder-llm-analyzer-bootstrap.html?appVersion=0.0.1

📖 Usage

1. Configure Your LLM

On first launch, you'll be prompted to configure your AI provider:

Your API key is stored locally in your browser and never sent anywhere except to your chosen LLM provider.

2. Select a Folder

Click "Choose Folder" and select a folder containing files you want to analyze.

3. Enter Your Query

Type your question or analysis request (e.g., "Summarize the main points", "What are the key findings?")

4. Start Analysis

Click "Start Analysis" to begin processing. You'll see real-time progress for each file.

5. Export Results

Download results as Excel (.xlsx) or CSV files.

📁 Supported File Types

  • Text Files: .txt, .md, .json, .csv, .xml, .html
  • Code Files: .js, .ts, .jsx, .tsx, .py, .java, .cpp, .c
  • Style Files: .css, .scss, .sass
  • Config Files: .yml, .yaml, .log, .sql
  • Documents: .pdf

💡 Example Queries

  • "Summarize the main points of each document"
  • "What are the key findings or conclusions?"
  • "Extract all dates mentioned in the files"
  • "What are the common themes across all documents?"
  • "Identify any security concerns or vulnerabilities"
  • "What are the technical requirements mentioned?"

🔧 Advanced Configuration

URL Parameters

Control the loading behavior with URL parameters:

?loader=@kosmas10/html-app-loader          # Use different Stage 2 loader
?loaderVersion=1.0.0                        # Pin Stage 2 version
?app=@kosmas10/folder-llm-analyzer          # Load different app
?appVersion=0.0.1                           # Pin app version
?appFile=folder-llm-analyzer.html           # Load specific file

Example: Pin All Versions

folder-llm-analyzer-bootstrap.html?loaderVersion=1.0.0&appVersion=0.0.1

📦 Package Structure

@kosmas10/folder-llm-analyzer/
├── folder-llm-analyzer.html              # Main application (Stage 3)
├── folder-llm-analyzer-bootstrap.html    # Bootstrap loader (Stage 1)
├── README.md                              # This file
└── LICENSE                                # MIT License

🌐 Browser Compatibility

  • ✅ Chrome/Chromium 90+
  • ✅ Firefox 88+
  • ✅ Safari 14+
  • ✅ Edge 90+

🔒 Privacy & Security

  • Local Storage: API keys stored in browser localStorage only
  • No Tracking: No analytics or tracking code
  • Direct API Calls: Your data goes directly to your chosen LLM provider
  • Open Source: Full source code available for inspection

🐛 Troubleshooting

Application Won't Load

  1. Check your internet connection
  2. Ensure cdn.jsdelivr.net is accessible (not blocked by firewall)
  3. Try clearing browser cache
  4. Check browser console for errors (F12)

LLM Not Responding

  1. Verify your API key is correct
  2. Check you have API credits/quota available
  3. Ensure the API endpoint is accessible
  4. Try reconfiguring with the LLM Config button

Folder Won't Upload

  1. Ensure your browser supports the File System Access API
  2. Try selecting a smaller folder first
  3. Check that files are in supported formats

🛠️ Technologies Used

  • Vanilla JavaScript (ES6+)
  • HTML5 & CSS3
  • Lucide Icons (CDN)
  • XLSX.js (Excel export)
  • Anthropic Claude API
  • Modern CSS Grid & Flexbox

📝 Version History

0.0.4 (Current)

  • Removed all non-error console output to prevent console popups during normal operation
  • Bootstrap file updated with cleaner code structure and version resolution from package.json
  • Cleaner console output - only shows errors when something goes wrong

0.0.3

  • Improved Claude Artifact detection using window.fs and window.storage APIs
  • More reliable detection method that checks for Claude-specific APIs instead of URL patterns
  • Better compatibility with Claude artifact environment

0.0.2

  • Added Claude Artifact detection and automatic configuration
  • Improved environment detection for seamless Claude integration

0.0.1

  • Initial release
  • Multi-format file support (text, code, PDF)
  • Batch processing with progress tracking
  • Excel and CSV export
  • 3-stage loading architecture

🤝 Contributing

Contributions are welcome! Please visit: https://github.com/kosmas10/ai-chat-extensions

📄 License

MIT License - See LICENSE file for details

🆘 Support

For issues, questions, or feature requests:

Made with ❤️ for AI-powered folder analysis