feat(slides): align deck workflow with frontend skill contracts#31
Merged
Conversation
evidence-root.mjs now accepts an optional lane (frontend default, slides allowed) so sibling skills share one run-root contract instead of hand-rolling paths; unknown lanes fail closed.
…acts - Style discovery gains a StyleGallery route: a user-named brand or mood loads exactly one teardown from the frontend brand library as scoped reference input (never automatic token authority), translated into 1920x1080 stage tokens with provenance recorded in VISUAL_QA.md. - The anti-slop reference now names the rewritten frontend contract and binds its non-waivable core (honest assets, no shipped placeholders, named waiver authority) to every deck. - Supplied deck imagery is authority-classified the image-first way; animation patterns declare the shared motion-core floor. - The visual-QA gate creates and verifies its run root through the shared portable evidence helper's slides lane. - Contract test pins the gallery scoping, anti-slop core, motion floor, and evidence-lane wiring.
Synchronize the npm, Codex, and Claude manifests for the slides alignment feature.
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
Express the shared evidence helper relative to the announced $SKILL_DIR instead of an unbound $FRONTEND_SKILL_DIR, tighten the gallery trigger to a user-named brand, pin every cross-skill reference target's existence in the contract test, make the unknown-lane rejection assert its actual error, and fold the slides alignment into the design-audit decision row.
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.
Summary
references/design/_INDEX.md) as scoped reference input — never automatic token authority — translated into 1920×1080 stage tokens, with the teardown named as provenance inVISUAL_QA.md; without a user-named brand the preset/bold-template mix stays the defaultprefers-reduced-motion) in the animation reference instead of carrying an unsynchronized copyevidence-root.mjs create <slug> slides→.superloopy/evidence/slides/<ts>-<slug>; unknown lanes fail closed) so both skills share one run-root contract, and route the slides visual-QA gate through itWhy
PR #29 substantially upgraded the frontend skill (StyleGallery, rewritten proportional anti-slop with a non-waivable core, motion-core/motion split, image-first authority classification, portable evidence-root helper). The slides skill — which explicitly claims to be "this skill's own instance of the superloopy-frontend rule set" — had inherited none of it: its anti-slop reference described a contract that no longer exists, its style discovery could not ground a brand-named request in any reference, its animation rules duplicated the motion floor, and its evidence paths were hand-rolled while a hardened shared helper sat one directory away.
Validation
npm test— 653/653 passednpm pack --dry-run --json— version 0.12.0; 507 filesdocs/superloopy-loop-golden-set.mdstays exactly at the 550-line reviewability cap (rows updated in place)test/frontend-routing-scenarios.test.js(slides-lane create/verify, unknown-lane rejection)