Import & Export
Lore DB supports importing PDFs and JSON files, and exporting libraries for backup and sharing. Your data is yours — import fast, export any time.
Importing a PDF
Click Import PDF in the sidebar (or go to /import/pdf). Drag and drop a PDF file or click to browse.
What happens:
- Text is extracted from the PDF entirely in your browser — the file is never uploaded to our servers
- A preview shows you what will be created before anything is saved
- Small PDFs (under ~5000 words) become a single document — you choose the title and library
- Large PDFs are automatically split into chapters based on detected headings (Chapter N, numbered sections, etc.) — each chapter becomes a separate document in a new library
- You can edit all titles before confirming
- Documents are created, indexed, and immediately searchable
Supported formats: Text-based PDFs only. Scanned image PDFs are not supported (no OCR). Password-protected PDFs must be unlocked first. Max file size: 50 MB.
PDF import is completely free — it uses no AI credits. Text extraction happens in your browser, and document indexing uses the same free processing as manually created documents.
Exporting a library as JSON
Open any library you’re a member of and click Export. This downloads a JSON file containing the library name, description, and all document titles and content. This is your backup — store it somewhere safe.
{
"name": "Engineering Docs",
"description": "Internal engineering documentation",
"documents": [
{
"title": "Deployment Runbook",
"content": "# Deployment Steps\n\n1. Run tests..."
},
{
"title": "API Reference",
"content": "## Endpoints\n\n- GET /api/docs..."
}
]
} Importing a library from JSON
On the Libraries page, click Import and select a JSON file in the format shown above. A new library is created with all the documents, and each document is automatically indexed for search. You become the owner of the imported library.
What’s included (and what’s not)
Exports include document titles and content. They do not include members, roles, embeddings, or metadata like creation dates. When you import, all documents are re-embedded from scratch, so the search quality is identical to manually created documents.
Importing a library generates embeddings for each document. For large libraries (50+ documents), consider adding your own OpenAI API key in Settings first.
Make a habit of exporting important libraries periodically. It only takes a click and gives you a portable backup you can import anywhere.