About GUNDAM Archive
A non-commercial fan research project. The goal is to turn scattered setting notes into something you can check, trace and keep building on.
Why this exists
Gundam spans dozens of eras and works. The facts live in anime, setting books, model kit manuals and community wikis. Numbers disagree, timelines do not line up, and comparing one tech line can mean reading a dozen works. GUNDAM Archive puts that material into structured data.
What is in the box
A static archive built with Next.js 15. All content lives in JSON under content/ and is checked by Zod at build time. The 3D viewer uses procedural reference assets. A glTF LOD0 can be wired in through the asset manifest without changing page code.
Content and code stay separate. Add a suit by adding a JSON file. No page edits.
One source of truth: src/lib/schema.ts defines both the data shape and the TypeScript types.
The build is the check. A bad schema, foreign key, lineage loop or one-way relation stops the release.
Three-language routes: /zh, /ja and /en, with an explicit fallback when a translation is missing.
Content pipeline
Adding a suit does not touch page code: drop in JSON, run the checks, rebuild.
Add or edit an entry in content/*.json
Zod checks fields, foreign keys and references
Static export generates zh / ja / en pages
CDN / GitHub Pages, cache busted by hash
src/lib/schema.ts defines both the data shape and the TypeScript types. Zod parses every file at build time, so a bad field, foreign key or lineage loop stops the build instead of shipping.
src/lib/schema.tsData rules
Sources are split into S/A/B/C tiers. Missing numbers stay blank. Values are never estimated from a similar suit. Disputed entries list both readings and carry a disputed flag. Every lineage edge needs a source. Events, people and tech entries need a long-form explanation. Public editing is closed; we take sourced correction drafts.
Roadmap
This release is a working archive, not the final form. Next up: real glTF assets and a content service.
M0 Foundation: repo, CI, design tokens, schema
DONEM1 One suit, end to end: 3D viewer, exploded view, quantum mode, dossier
DONEM2 Data skeleton: 12 suits, timeline, lineage, trilingual shell
DONEM3 Relations: people, factions, maps, search, motion system
WIPM4 Scale: 147 suits, global lineage graph, performance pass
NEXTM5 Launch: accessibility, mobile, SEO, monitoring, public beta
NEXTHow to help
Corrections need a source and page number. For a new suit, read docs/CONTENT_GUIDE.md and run npm run validate before opening a PR. Code follows Conventional Commits.
GitHub Repositorydocs/CHRONOLOGY.mdTwelve-era table and reading pathsdocs/MACHINE_COVERAGE.mdSuit coverage and inclusion rulesdocs/ARCHITECTURE.mdArchitecture and decision recorddocs/CONTENT_GUIDE.mdHow to add a suit, event or characterdocs/DATA_DICTIONARY.mdData dictionary and unitsdocs/ASSET_PIPELINE.mdglTF / LOD / manifest rulesdocs/GOVERNANCE.mdSource tiers, disputes and copyright
Copyright and licence
Code is MIT. Original text and charts are CC BY-NC-SA 4.0. Gundam and all related names and settings belong to Sotsu・Sunrise and Bandai Namco. This is an unofficial fan project. It uses no official artwork and is not sold.
Unofficial fan project. Not affiliated with Bandai Namco or Sotsu・Sunrise.