Legends of Clal

Josh Holladay's fantasy series, written as J. A. Holladay. Book 1: The Dragon Summoner.

Consolidated 2026-10-23 from the workspace folder Legends of Clal/ plus related files from elsewhere on the VPS. Copies only; originals were left where they were.

Path	What it is
`legends-claude.md`, `legends-log.md`, `legends-history.md`	Project context, append-only decision log, history
`Legends of Clal - AI/`	AI-era working files: world bible, continuity index, scene outline
`legends of clal hub - july 22 2026/`	The Hub: 2022 master drafts, Clal 2014 archive, world building, techniques
`july2026-novelcrafter-import/`	NovelCrafter export (characters, locations)
`Mentors and Coaches/`	Mentor material
`drive-import/`	Files pulled from Google Drive that weren't already here (the 8 Book Plots, Book 2 draft, Book 1 Version 4)
`text-extracts/`	Plain text pulled out of old `.doc`, `.pdf`, `.pages`, `.numbers` and `.xlsx` files so they can be read and searched
`foundations/`	The World of Clal: Foundations (Oct 10, 2007), the original worldbuilding notes, verbatim
`references/four-waters/`	Fountain / Four Waters diagram (redraw; original image to be added)
`from-switch26/`	The Switch26 project brief page and its copy of `legends-claude.md`
`from-aios-commands/`	The `/legends-prime` command from the AIOS workspace

The explorer site

Open site/index.html (or the index.html at the repo root, which forwards to it) in any browser. Or run ./open-site.sh and go to http://localhost:4880/site/index.html.

Keep the site/ folder inside the repo. Its pages link to the original files with ../ paths, so if site/ is copied somewhere on its own, or opened from a server whose root is site/, the "original" links, images, PDFs and audio will 404. GitHub's web view won't render it either; use a local clone. It is a card-based way to explore everything here: a front door of curated cards (foundations, world, story, characters, locations, process), a searchable Archive of every file, a picture Gallery, and Audio.

The site does not copy the originals. Reading pages are small HTML files in site/p/, and every "original" link points back to the real file in this repo.

Rebuild after adding or changing files:

pip install markdown mammoth olefile pypdf openpyxl numbers-parser cramjam   # once
python3 extract_text.py          # only when old-format files change (about 4 minutes)
python3 build_site.py            # rewrites site/ in about a minute

Notes: every text-like file becomes a readable page (Markdown, .docx, plus .doc, .pdf, .pages, .numbers and .xlsx via text-extracts/). Pages-file text is a best-effort recovery and the original is always linked. Four old .doc files (Called to the Journey, Called to the Journey 2, Sedrak Illia, The Rishk The Banished) contain no extractable text (likely pictures or fonts). Pictures inside Word files are not carried onto the reading pages. To change the front-door cards, edit build_index() in build_site.py.