rbrooks/Quest-Board

Watch

Fork 0

docs(design): catalogue every surface in Quest Board for the mockup round (#657) #658

Merged

claude-bot merged 1 commit from docs/657-surface-catalogue into main

2026-09-15 19:47:53 +00:00

claude-bot commented

2026-09-15 19:30:09 +00:00

Collaborator

Closes #657.

Why

The owner decided that nothing gets built until every screen and surface in Quest Board has been mocked up, for players and for GMs, and reviewed (#535, #656). This PR adds the list those mockups work from: every page and surface anyone can see, with every current and planned feature placed on one. It exists so that nothing is dropped or forgotten in the redesign.

It is a list, not a verdict. It does not check that features work, and it changes no code.

What's in it

docs/design/redesign-surfaces/:

CATALOGUE.md: 101 surfaces, organised by Design A (Book & Table) with the six shell decisions accepted on #656.
- Every surface opens with a summary: who sees it, devices, where it's reached from, whether Design A draws it, the states a mockup must show, what it opens, and what to watch for.
- Its features are split into three tables: built today (with file and line), built on the server with no screen yet, and planned (with issue).
- It starts with thirteen decisions for the owner, each with a proposal.

ledgers/: the ten read-only sweeps it was built from, 1,566 rows in all. Each ledger lists the files the sweep read, the containers it found, and its notes for the designer.

Sweep	Area	Rows
SH	shell, sign-in, invite, help, public analytics, Home, Profile	167
CA	campaign page and panels, Highlights, Analytics	243
SE	session page	157
TB	Shelf and Table view	112
PR	Prep and all thirteen tools	158
WK	wiki	199
AD	admin and content packs	151
EX	Discord bot, email, calendar, installed app, backend-served pages	55
API	server endpoints no screen calls	27
IS	open issues and design records	297

sources/: the surface list, placement decisions, summaries and build.ps1. The build fails if any row lands on no known surface, or if a surface has no summary, and the catalogue ends with a generated coverage check (1,566 of 1,566 placed).

Findings worth knowing before the mockups

Built on the server, with no screen:
- a wiki trash with restore
- the whole rightsholder-notice and takedown workflow
- instance health (undecryptable secrets, AI queue)
- the admin campaign list and owner reassignment
- content-pack search and browse
- summarisation run history
- per-speaker transcript stats
- the per-session character pin (#422)
The catalogue gives each a surface.
Four surfaces were added to the plan the sweeps started from:
- PUB-9: the Discord-linking pages the backend serves outside the React app
- WIKI-18: deleted articles
- ADMIN-13: content notices and takedowns
- ADMIN-14: instance health
Two more are proposals: PUB-10, an accessibility statement (decision D4), and DEV-1, the design-system reference #364 asks for.
The session sweep suspected the full transcript might reach players. Checked: it doesn't, because the server withholds it from anyone who isn't a GM (session_service.py:277). The appendix records the correction.

Checked

Built on Windows PowerShell 5.1.
Every feature table has five columns, and all 101 index links resolve.
The ledgers, summaries and intro contain no local paths or secrets.
CI has no docs checks; this PR changes no code.

🤖 Generated with Claude Code

Closes #657. ## Why The owner decided that nothing gets built until every screen and surface in Quest Board has been mocked up, for players and for GMs, and reviewed (#535, #656). This PR adds the list those mockups work from: **every page and surface anyone can see, with every current and planned feature placed on one.** It exists so that nothing is dropped or forgotten in the redesign. It is a list, not a verdict. It does not check that features work, and it changes no code. ## What's in it `docs/design/redesign-surfaces/`: - **`CATALOGUE.md`:** 101 surfaces, organised by Design A (Book & Table) with the six shell decisions accepted on #656. - **Every surface opens with a summary:** who sees it, devices, where it's reached from, whether Design A draws it, the states a mockup must show, what it opens, and what to watch for. - **Its features are split into three tables:** built today (with file and line), built on the server with no screen yet, and planned (with issue). - **It starts with thirteen decisions for the owner**, each with a proposal. - **`ledgers/`:** the ten read-only sweeps it was built from, 1,566 rows in all. Each ledger lists the files the sweep read, the containers it found, and its notes for the designer. | Sweep | Area | Rows | |---|---|---:| | SH | shell, sign-in, invite, help, public analytics, Home, Profile | 167 | | CA | campaign page and panels, Highlights, Analytics | 243 | | SE | session page | 157 | | TB | Shelf and Table view | 112 | | PR | Prep and all thirteen tools | 158 | | WK | wiki | 199 | | AD | admin and content packs | 151 | | EX | Discord bot, email, calendar, installed app, backend-served pages | 55 | | API | server endpoints no screen calls | 27 | | IS | open issues and design records | 297 | - **`sources/`:** the surface list, placement decisions, summaries and `build.ps1`. **The build fails if any row lands on no known surface, or if a surface has no summary**, and the catalogue ends with a generated coverage check (1,566 of 1,566 placed). ## Findings worth knowing before the mockups - **Built on the server, with no screen:** - a wiki trash with restore - the whole rightsholder-notice and takedown workflow - instance health (undecryptable secrets, AI queue) - the admin campaign list and owner reassignment - content-pack search and browse - summarisation run history - per-speaker transcript stats - the per-session character pin (#422) The catalogue gives each a surface. - **Four surfaces were added** to the plan the sweeps started from: - PUB-9: the Discord-linking pages the backend serves outside the React app - WIKI-18: deleted articles - ADMIN-13: content notices and takedowns - ADMIN-14: instance health **Two more are proposals:** PUB-10, an accessibility statement (decision D4), and DEV-1, the design-system reference #364 asks for. - **The session sweep suspected the full transcript might reach players.** Checked: it doesn't, because the server withholds it from anyone who isn't a GM (`session_service.py:277`). The appendix records the correction. ## Checked - Built on Windows PowerShell 5.1. - Every feature table has five columns, and all 101 index links resolve. - The ledgers, summaries and intro contain no local paths or secrets. - CI has no docs checks; this PR changes no code. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-bot added 1 commit

2026-09-15 19:30:10 +00:00

docs(design): catalogue every surface in Quest Board for the mockup round (#657 )

CI / Summarisation accuracy eval harness (stub provider) (pull_request) Successful in 58s

Details

CI / Bot/backend version sync (pull_request) Successful in 54s

Details

CI / Docker image build (pull_request) Successful in 26s

Details

CI / Backend lint (ruff) (pull_request) Successful in 1m2s

Details

CI / Frontend tests, audit, and build (pull_request) Successful in 2m53s

Details

CI / Bot tests and audit (pull_request) Successful in 2m43s

Details

CI / Backend migration, tests, and audit (pull_request) Successful in 8m55s

Details

CI / Synthetic session harness (no GPU, no LLM) (pull_request) Successful in 15m54s

Details

6e719a9662

The owner decided nothing gets built until every screen and surface has a
mockup, for players and GMs, and has been reviewed. This is the list those
mockups work from: every page and surface anyone can see, and every feature,
current or planned, placed on one of them, so nothing is dropped in the
redesign.

- CATALOGUE.md: 101 surfaces organised by Design A (Book & Table), with the
  shell decisions accepted on #656. Each surface has a summary (who, devices,
  reached from, whether Design A draws it, states to mock, what it opens, what
  to watch for), and its features split into built today, built on the server
  with no screen, and planned. It opens with thirteen decisions for the owner.
- ledgers/: the ten read-only sweeps it was built from, 1,566 rows in all:
  - shell and public pages, and Home
  - campaign, session, Shelf and Table, Prep, wiki
  - admin and content packs
  - outside the web app
  - server capabilities with no screen
  - the open issues and design records
- sources/: the surface list, placement decisions, summaries and the build
  script. The build fails if any row lands on no known surface, or if a
  surface has no summary.

The sweeps found capabilities with no screen, which the redesign now has a
place for: a wiki trash, the rightsholder-notice workflow, instance health,
the admin campaign list, and content-pack search.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

claude-bot scheduled this pull request to auto merge when all checks succeed

2026-09-15 19:30:25 +00:00

claude-bot merged commit 0986bf41b7 into main

2026-09-15 19:47:53 +00:00

claude-bot deleted branch docs/657-surface-catalogue

2026-09-15 19:47:54 +00:00

claude-bot referenced this pull request from a commit

2026-09-15 19:47:55 +00:00

Merge PR #658: catalogue every surface in Quest Board for the mockup round (#657)

claude-bot referenced this pull request

2026-09-15 19:48:59 +00:00

[UX] Mock up every catalogued surface, for players and GMs, before building #659