# Dicta Genizah Search

> A collaborative research platform for the Cairo Genizah: full-text
> search across ~255,000 manuscript fragments, with transcriptions,
> high-resolution images, scholarly metadata, parallels, and
> midrash/halakhic source linking. Bilingual (Hebrew / English).

Maintained by Dicta — Israel Center for Text Analysis. Public,
non-commercial scholarly resource.

## Primary entry points

- [Home](https://genizahsearch.com/): site overview and navigation
- [About](https://genizahsearch.com/about): project background and the Cairo Genizah
- [Help](https://genizahsearch.com/help): user guide and search tips
- [Catalog browse](https://genizahsearch.com/catalog-browse): structured browse by domain, author, work
- [Browse manuscript](https://genizahsearch.com/browse): per-manuscript view (e.g. /browse?sys_id=...)
- [Discoveries](https://genizahsearch.com/discoveries): community-shared joins and findings
- [Download desktop](https://genizahsearch.com/download): Windows desktop application

## Machine-readable endpoints

- [OpenAPI](https://genizahsearch.com/api/openapi.json): public search API contract
- [API docs](https://genizahsearch.com/api/docs): Swagger UI
- [Sitemap index](https://genizahsearch.com/sitemap.xml): all indexable URLs

## What is the Cairo Genizah

The Cairo Genizah is a collection of ~400,000 Jewish manuscript
fragments discovered in the Ben Ezra Synagogue in Fustat (Old Cairo).
It spans roughly the 9th-19th centuries and includes biblical,
rabbinic, liturgical, halakhic, mystical, scientific, and documentary
material in Hebrew, Aramaic, Judeo-Arabic, and other languages.
Major holding libraries include Cambridge University Library (Taylor-Schechter),
the Jewish Theological Seminary, the National Library of Russia,
Oxford Bodleian, the British Library, and the National Library of Israel.

## Crawler guidance

- /search and /parallels are dynamic result pages — prefer /browse?sys_id=...
  for stable, citation-worthy manuscript URLs.
- Per-manuscript metadata (shelfmark, library, transcription, bibliography)
  is rendered client-side via WebSocket hydration; allow time for the
  full DOM before extraction.
- User-account, admin, and API paths are excluded — see /robots.txt.