Repository navigation
Rewrite the lobby trivia example as a site guide - #1452
Merged
Merged
Conversation
docs/product/lobby-trivia-3player-external-public.md was the one worked example in docs/product/, written in caps emphasis with #164, #221 and #304 references, and it said true phone widths were a future upgrade. mobileEmulation has shipped since, so the caveat described a limit the study can now remove. site/content/docs/public-lobby.mdx, "Run three phone participants in one public lobby", replaces it: - what the study does, in plain sentences; - "Make the participants phone-sized": the 500 CSS px floor without emulation, then the execution.desktop.fidelity.mobileEmulation block and what it gives, linking /docs/computer-use#mobile-emulation; - run commands with npx, the dry run first; watching from a phone uses serve with the real flag spellings (--tunnel ngrok --oauth google --allow-email), since watch --expose streams only computer-use runs; - what the evidence claims, as a list; - the two app details to recheck before a live run. The sidebar lists it under Guides, and docs/README.md drops the old entry. Commands and paths keep "lab" until the study rename lands. Checked: format, lint, knip, docs:check, prose:check, vocabulary:check and typecheck pass; tests/surface passes; site typecheck, build and registry:check pass, and the built page has its H1 and the mobile-emulation link. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
danielgwilson
added a commit
that referenced
this pull request
Oct 3, 2026
prose:check now reads the current docs under three roots, each capped in scripts/caps.json at prose.<root>.<kind>: - docs: the root guides (README, AGENTS, ARCHITECTURE, CONTEXT, CONTRIBUTING, SECURITY, TELEMETRY) and docs/ outside history/ and evidence/. - site: site/content/docs/*.mdx except the generated cli.mdx. - evidence: docs/evidence/. Dated study records are capped so a new record adds no tells, and are not cleaned, since a record keeps the words it was written in. docs/history/ and CHANGELOG.md are not read. Kinds: issue-refs, caps, em-dashes, invariant-refs, authority, honest, archaeology (plus SLICE 2, phase 2, PR1 of) and contrast (not just, not merely, rather than). scripts/lib/doc-prose.mjs blanks front matter, fenced code, code spans, link targets, URLs and markup tags before counting, and reads repo file names and acronyms (README, WAI-ARIA) as names in the caps count. Caps are the counts on 9f033af, after the docs cleanups #1446, #1452, #1454, #1455, #1462, #1463, #1471, #1472 and #1476, in the kind order above: docs 1/0/0/0/0/1/0/20, site 0/0/0/0/0/0/0/11, evidence 102/64/31/1/3/5/2/11. The docs issue-ref and honest at 1 are the last hits, which #1486 removes. The 20 docs and 11 site contrast hits each name two alternatives a reader would weigh. AGENTS.md and CONTRIBUTING.md say prose:check covers docs. Checked on 9f033af: format:check, lint, knip, docs:check, prose:check, vocabulary:check, typecheck, tests/scripts (211 tests, 5 new in check-doc-prose.test.ts), check-cap-direction against origin/main (new caps only). Not checked: the full vitest suite. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
danielgwilson
added a commit
that referenced
this pull request
Oct 3, 2026
prose:check now reads the current docs under three roots, each capped in scripts/caps.json at prose.<root>.<kind>: - docs: the root guides (README, AGENTS, ARCHITECTURE, CONTEXT, CONTRIBUTING, SECURITY, TELEMETRY) and docs/ outside history/ and evidence/. - site: site/content/docs/*.mdx except the generated cli.mdx. - evidence: docs/evidence/. Dated study records are capped so a new record adds no tells, and are not cleaned, since a record keeps the words it was written in. docs/history/ and CHANGELOG.md are not read. Kinds: issue-refs, caps, em-dashes, invariant-refs, authority, honest, archaeology (plus SLICE 2, phase 2, PR1 of) and contrast (not just, not merely, rather than). scripts/lib/doc-prose.mjs blanks front matter, fenced code, code spans, link targets, URLs and markup tags before counting, and reads repo file names and acronyms (README, WAI-ARIA) as names in the caps count. Caps are the counts on 9983746, after the docs cleanups #1446, #1452, #1454, #1455, #1462, #1463, #1471, #1472, #1476 and #1486, in the kind order above: docs 0/0/0/0/0/0/0/20, site 0/0/0/0/0/0/0/11, evidence 102/64/31/1/3/5/2/11. The 20 docs and 11 site contrast hits each name two alternatives a reader would weigh. AGENTS.md and CONTRIBUTING.md say prose:check covers docs. Checked on 9983746: format:check, lint, knip, docs:check, prose:check, vocabulary:check, typecheck, tests/scripts (231 tests, 5 new in check-doc-prose.test.ts), check-cap-direction against origin/main (new caps only). Not checked: the full vitest suite. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Beauty pass, package 8 (docs-restructure-and-package-files), PR 8c, second half (the first half is #1446). The plan is
handoffs/2026-10-02-beauty/docs-restructure-plan.md, step 8.What changed
docs/product/lobby-trivia-3player-external-public.mdwas the only worked example indocs/product/. It used caps emphasis ("ONE public multiplayer lobby", "EXTERNAL-PUBLIC", "NO touch input"), cited #164, #221 and #304, and called true phone widths "the #221 CDP-device-emulation upgrade". That setting has shipped since:execution.desktop.fidelity.mobileEmulation.site/content/docs/public-lobby.mdx, "Run three phone participants in one public lobby", replaces it:/lobby/CODEas the pass signal.mobileEmulationblock and what it gives (preset width, device pixel ratio, touch, mobile user agent, read-back underdesktopGeometry.fidelity). It links/docs/computer-use#mobile-emulation.npx, dry run first. Watching from a phone usesservewith real flag spellings:--tunnel ngrok --oauth google --allow-email. The old page had--allow-emailsand called--exposeon the concurrent path a "0.20.1 fast-follow";watch --exposestreams only computer-use runs.The sidebar lists it under Guides, after computer-use, and
docs/README.mddrops the old entry. Commands and paths keep "lab" until the study rename lands.No release line: docs only.
Checked
tests/surfacepasses.registry:checkpass. The built page has its H1 and the/docs/computer-use#mobile-emulationlink.Not checked
mobileEmulation. The emulation facts come fromcomputer-use.mdxandexternal-public-shared-world.md.humanish/labs/lobby-trivia-3player.yamlkeeps its own caps text and caveat. The starter-files package (contracts: define run bundle, adapter, actor, substrate, evidence, and review schemas #4) owns lab text.🤖 Generated with Claude Code