FileHawk - Local Semantic File Search Tool

A powerful desktop application that uses semantic search to find files based on their content, not just filenames. Built with modern web technologies and powered by advanced AI models.

Features

• Semantic Search: Find files using natural language queries instead of exact keywords
• Multiple File Types: Supports PDF, DOCX, TXT, MD, XLSX, CSV, Python, JavaScript, and many more
• Advanced AI Models: Uses sentence transformers for deep content understanding
• Modern Desktop UI: Clean, responsive interface built with React and TailwindCSS
• Native Integration: System file dialogs and cross-platform compatibility
• Real-time Progress: Live indexing progress with detailed file tracking
• Smart Chunking: Intelligent content splitting for optimal search results
• Background Processing: Index large folders without blocking the interface

Installation

For detailed setup instructions, see Setup_Instructions.md Settings are automatically saved and persist between sessions.

Supported File Types

Text Files

• .txt, .md, .rst, .tex
• .py, .js, .ts, .java, .cpp, .c, .h
• .html, .css, .xml, .json, .yaml, .yml
• .sh, .bash, .zsh, .fish

Documents

• .pdf (using pdfplumber)
• .docx (using python-docx)
• .pptx (using python-pptx)
• .rtf (using striprtf)
• .odt, .ods, .odp (using odfpy)

Data Files

• .csv (using pandas)
• .xlsx (using openpyxl/pandas)
• .json (built-in)

Other Formats

• .html (using BeautifulSoup)
• .epub (using ebooklib)
• .eml (using email module)
• .msg (using extract-msg)

Architecture

FileHawk uses a hybrid architecture combining modern web technologies with powerful AI:

• Electron Frontend: Cross-platform desktop app built with React and TypeScript
• Python Backend: Flask API server handling file processing and AI operations
• ChromaDB: High-performance vector database for semantic search
• SentenceTransformers: State-of-the-art AI models for content understanding
• Secure IPC: Safe communication between frontend and backend processes

Troubleshooting

Common Issues

Installation Problems

• Verify Python 3.8+ and Node.js 16+ are installed
• Run pip install -r requirements.txt to install dependencies
• Make sure npm install completed without errors

Application Won't Start

• Check if port 5001 is available (backend uses this port)
• Ensure Python backend can access the selected folders
• Try running npm run dev from the desktop-app directory

Search Not Working

• Make sure folders are indexed first (look for "Index Complete" status)
• Try different search terms or check file permissions
• Verify indexed folders still exist and are accessible

For detailed troubleshooting, see Setup_Instructions.md

License

MIT License - see LICENSE file for details.

Acknowledgments

• SentenceTransformers for semantic embeddings
• ChromaDB for vector storage
• Electron for cross-platform desktop apps
• React and TailwindCSS for the UI