Excavate
Intelligent search across thousands of PDF pages
What is Excavate?
Excavate is a desktop application for managing and searching through PDF documents with advanced text extraction and bookmarking capabilities.
It is also an experiment in vibe-coding, to see how far I can take building a complex application with a focus on user experience, without any manual writing of code. The entire application, including the README, is generated through prompts and interactions with AI tools. This approach allows for rapid iteration and a unique development process, but may also lead to some quirks or inconsistencies as the AI learns and adapts, as well as high api/token costs.
Features
- Intelligent search across thousands of PDF pages
- Text extraction using OCR on your device
- Bookmarking capabilities for quick access to important pages
- Real-time file system monitoring
- Local storage: all data kept in
documents/Excavate(settings, Python environment, saved searches) and within your library directories (processed indexes)
Why was it built?
Singapore's building code documents are overwhelmingly in PDF format, which makes searching for specific information rather like looking for a needle in a particularly bureaucratic haystack. I built Excavate as a search tool capable of searching through multiple PDF documents with ease. It saves time, certainly, but more importantly, it preserves one's sanity when the alternative is another afternoon lost to endless scrolling.
Screenshots
