Benchmark realistic bolder and polish tasks across fast, full-model, and full-context worker profiles with deterministic and independent quality gates.\n\nAI-assisted: OpenAI Codex.
Introduce a Live-owned app-server supervisor with progressive fenced publishing, partitioned control polling, cancellation and recovery safety, and measured integration coverage.
AI-assisted implementation under maintainer direction.
Four gaps from human review of gpt-5.6 eval artifacts:
1. codex-grid-background variants: the block scan now also matches the
inverted end-of-tile hairline form (transparent calc(100% - Npx))
and reads the tile cell from the background shorthand's `/ Npx Npx`
slot, not just background-size declarations. A single hairline layer
qualifies when tiled by a px pair cell (page-scale line field);
percent-tiled single hairlines (background-size: 25% 100% rules on
data-viz tracks/graphs) stay legal.
2. hero-eyebrow-chip branch C (dash-prefix): sentence-case, regular-
weight microlabels above the h1 announced by a short chromatic
::before/::after bar (8-80px x 1-6px, accent fill). Static cascade
marks dash-pseudo targets during rule collection; the browser path
reads getComputedStyle(el, '::before'/'::after').
3. New `marquee` slop rule: <marquee> elements, and infinite animations
bound to keyframes with >= 20 percentage points of X travel. Percent
travel only — px-travel loops are bespoke product animations
(waveform playheads, progress sweeps). Centered elements animating
other properties (constant -50% X), non-infinite slide-ins, rotations,
and pulses never qualify.
4. side-tab inset box-shadow variant: single-edge inset shadows
(3-12px offset on one axis, no blur/spread, chromatic) drawn as
stripes on cards/badges/menu items. Selection-state indicators
([aria-current], [aria-selected], [role=tab], active/current/selected
hints, interaction states) stay exempt; the same stripe repeated
unconditionally on every item flags. Narrow fixed-width glyphs
(logo marks) are exempt. isTabContextElement narrowed to match:
bare nav ancestry no longer blanket-exempts top/bottom border
stripes — only explicit tab semantics or state markers do.
Browser bundle regenerated.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Four changes driven by human design review of eval artifacts:
1. Static engine contrast fidelity (nav-CTA cascade miss):
- parseAnyColor evaluates color-mix() (premultiplied sRGB mix; exact
for the dominant `color-mix(in oklab, C n%, transparent)` chip form)
- extractStaticColor captures color-mix() balanced instead of plucking
"transparent" out of the expression
- resolveBackground composites translucent layers over the opaque base
in both engines instead of skipping (static) or returning them
as-if-opaque (browser)
- NEW hover pass in the static cascade: :hover rules are matched via
state-stripped selectors, merged per-property against the resting
cascade with real specificity, and checked for WCAG contrast on
styled controls (checkHoverContrast). Catches the recurring miss
where a broader selector (.nav-links a:hover) beats the CTA's own
hover color and drops the pair below AA.
2. New `radial-halo` slop rule: chromatic radial-gradient wash (visible
saturated center -> transparent) as a decorative background on a dark
page. Exempts achromatic vignettes, opaque-end sheens, px-stop dot
textures, url() photo layers, and translucent (<0.7 alpha) staged-
light washes. Separate id from dark-glow so dashboards track the
gradient-drawn variant independently.
3. side-tab horizontal variant: 3-12px chromatic border-top/bottom (and
top/bottom-anchored full-width pseudo stripes) on cards/badges flag as
side-tab. Exempt: tablist/nav/aria-selected underlines, link/button
affordances, table cells, hr, state-conditional pseudo stripes, and
>12px bands. Badge-shaped spans (own visible background) participate.
4. CLI: file:// URLs route to the Puppeteer browser engine (~2s on a
50KB page), and detect --json findings now carry the registry
`category` field so downstream QA loops can separate mechanical slop
tells from judgment calls.
Fixture policy update: flat 3px top-accent cards moved from should-pass
to flag columns; tablist-underline and 16px-band pass cases added.
Browser bundle regenerated.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Measure framework and provider latency, enforce fidelity and cleanup gates, and publish reproducible results on the dev-only Live Lab.\n\nAI-assisted: OpenAI Codex.
Two gaps surfaced by human eval review of real artifacts:
1. side-tab missed the pseudo-element variant. The accent stripe drawn as
an absolutely-positioned ::before/::after (left/right: 0, top+bottom: 0
or height: 100%, narrow width, colored background) uses no border
property at all, so neither the element-level border checks (pseudo
elements never enter the static cascade or DOM walk) nor the
border-left/right regexes could see it. New scanCssTextForPseudoStripe
scans stylesheet text for that shape, mirroring the border rule's
gates: >= 3px thick (<= 12px), chromatic fill (var()-resolved, neutral
dividers skipped), full height against a side edge, with the
blockquote/prose exemptions preserved.
2. New pulsing-dot rule (slop): small circular "live" indicator dots
(<= 16px, border-radius >= 40% or pill values) bound to an infinite
animation whose keyframes vary opacity, scale, or box-shadow — or
pulse/blink/ping names when the keyframes aren't in the scanned text —
plus the Tailwind animate-ping/pulse + rounded-full + tiny-size utility
combo. Rotation-only keyframes (spinners) never flag, including when
they hide behind a pulse-like name.
Both scanners live in checkHtmlPatterns, so the static-html engine and
the browser bundle share the same detection path. Browser/extension
bundles regenerated; docs rule count bumped to 47.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Answering the obvious question the family-depth framing dodged: with
modes derived per task, files named for the old two-register taxonomy
had no architectural reason to exist. brand.md's surviving depth (lane
test + inverse test, reflex-reject lanes, color discipline, layout
moves, permissions) folds into new-work.md, where all of it belonged:
it is new-identity Persuade/Experience guidance. product.md's content
moves unchanged to operate.md, its true name. Both old files remain as
one-line redirect stubs because register: brand|product in existing
PRODUCT.md files and older links point there. All cross-references
retargeted (SKILL.md modes intro, context.mjs REGISTER hint, live.md,
typeset.md); 85 tests green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Architecture per Paul: impeccable is primarily a daily driver on
existing codebases; the always-loaded core should serve that 90% path,
not carry the full generative arsenal on every invocation. SKILL.md now
holds brief-wins, existing-worlds (the headline path), the four visitor
modes, the full craft floor, and a hard gate: new identity work
(greenfield, or a redesign discarding the current look) MUST read
reference/new-work.md before any design decision. That file carries the
generative playbook (seed, subject grounding, plan/self-check/signature,
hero-thesis, everything-bold, prove-don't-claim, color commitment,
calibration, persuade type/imagery). context.mjs enforces the gate
mechanically: NEW_WORK directive when no PRODUCT.md/DESIGN.md exists,
and the old mandatory register-file read is replaced by a REGISTER
family hint. No surfaces: map anywhere; mode is derived per task.
Gate compliance is measurable via skillEvidence.directSkillFileReads.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Eval evidence showed the per-edit PostToolUse stream fires overwhelmingly
on copy-level rules (em-dash-overuse ~97x/session) and measurably makes
models more conservative, while a full-detector pass at completion is what
actually fixes contrast/padding/glow. Split the hook accordingly:
- Per-edit (PostToolUse) now surfaces only IMMEDIATE_TIER_RULES: broken
output (broken-image, text-overflow, clipped-overflow-container,
body-text-viewport-edge), objective contrast/legibility failures
(low-contrast, gray-on-color, tiny-text), single-property mechanical
slop (gradient-text, dark-glow), and design-system drift (the four
design-system-* rules, which compound if left uncorrected). Everything
else defers. Override with hook.perEditRules: "all" in
.impeccable/config.json. Tiering is off for Cursor/Copilot harnesses,
which have no Stop pass wired, so nothing gets silently dropped there.
- Stop deep pass (runStopHook): runs the FULL rule set over every UI file
touched this session (tracked via the existing hook.cache.json session
state; deferred-only edits now mark the file touched), dedupes against
everything already surfaced per-edit, honors ignore-rule/file/value and
inline disables, reuses the [impeccable@1] envelope, and no-ops fast
when no UI files were touched. Emits hookSpecificOutput
{ hookEventName: "Stop", additionalContext } per the Claude Code SDK
Stop contract (conversation continues so the model can act on it).
Second Stop fire is silent - deep-pass findings are remembered.
- Wiring: Stop entries (timeout 30) in plugin/hooks/hooks.json, the
.claude settings + .codex hooks manifests (transformers + hook-admin
repair path). Claude Code and Codex both dispatch a native Stop event;
Cursor's stop hook is inconsistently dispatched (pre-write gate stays)
and Copilot's agentStop/sessionEnd don't inject model context, so
neither gets a Stop entry - documented in reference/hooks.md.
- Tests: tiering split/override/harness gating, Stop dedupe + silent
no-touched-files + ignore machinery + kill switches; existing per-edit
tests moved to immediate-tier rule ids. 181 tests green; smoke-tested
the built dist skill end to end (glow surfaced per-edit, em-dash only
at Stop, second Stop silent).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- parseAnyColor now covers oklab(), hsl()/hsla(), hwb(), and ~35 common
named colors on top of rgb/rgba/hex/oklch, so checkGlow sees the color
regardless of authoring format (Chrome preserves oklch() in computed
styles, which the old rgba-only match silently passed).
- checkGlow gains a second tell: a zero-offset chromatic box/text-shadow
with blur > 4px is flagged on ANY background (the halo pattern);
achromatic zero-offset shadows and focus rings stay legal. The
existing chromatic-blur-on-dark-background rule is unchanged in
semantics but now parses every color format.
- text-shadow is checked wherever box-shadow was (browser DOM path with
inherited-value dedupe, static engine via new textShadow cascade
support, text engines).
- The page-level text scan (regex engine + checkHtmlPatterns) is now a
shared scanCssTextForGlow that resolves single-level var() refs
against custom properties collected from the same text; unresolvable
var() in a shadow color position is skipped, never guessed. Its
dark-page heuristic accepts var()/oklch backgrounds but only when
declared at root scope (body/html/:root or body inline style).
- dark-glow keeps its id; registry name/description updated to cover
both cases.
Validated: three eval repro samples with oklch / var(--x) glows that
previously produced zero findings now flag on the static CLI path; ten
known-good largerun samples stay clean except one with genuine amber
status-dot halos (0 0 12px oklch(.73 .17 65/.4)).
Note: cli/engine/detect-antipatterns-browser.js and the extension
detector are generated and still need 'node scripts/build-browser-detector.js'
+ 'node scripts/build-extension.js' once builds are unblocked.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Add OpenAI plugin submission bundle
Build a Codex-native OpenAI plugin with bundled hooks, public listing metadata, submission guidance, privacy coverage, and regression tests.
AI assistance: OpenAI Codex prepared and validated these changes under maintainer direction.
* Fix provider script command rendering
Replace heuristic rewrites across executable scripts with one explicit provider marker, render pinned shortcuts per target harness, and remove the personal email from the public publisher manifest.
Addresses automated review feedback on PR #363.
AI assistance: OpenAI Codex prepared and validated these changes under maintainer direction.
* Add mechanical pre-scan for typeset and layout commands.
Introduce --scope filtering, layout/type rule scopes, DESIGN.md font-size validation, and pre-scan steps in the skill references so agents run detect before LLM judgment.
Fixes#149
Co-authored-by: Cursor <cursoragent@cursor.com>
* Add isolated sub-agent orchestration for typeset and layout pre-scans.
Run the mechanical detector and visual assessment in parallel sub-agents so deterministic findings cannot anchor LLM judgment, matching the critique pattern Paul requested on PR #345.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix: reject bare --scope so detect never scans unscoped by mistake.
When --scope had no value, the CLI dropped the flag and ran a full scan instead of failing, which could silently use the wrong rule set during typeset/layout pre-scans.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix: require both typeset and layout assessments in sub-agents.
Close a loophole where agents ran only the mechanical pre-scan inline by interpreting "running both" as permitting one inline assessment.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Abdul Wahab <abdulwahab@Abduls-MacBook-Pro-2.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
* Route native projects to native command variants for audit and adapt
Follow-up to #269. The web audit.md and adapt.md carried "translate this
yourself" Platform notes, so a native invocation paid for the full web
file (~1.8k / ~2.6k tokens, mostly inapplicable) and did error-prone
run-time translation. Authored with AI assistance (Claude Code) under
maintainer direction.
- New reference/audit.native.md and reference/adapt.native.md: authored
native content (VoiceOver/TalkBack, platform conformance, adaptivity
dimensions; phone-to-tablet, platform-to-platform, web-to-native
strategies). One variant per command covers ios, android, and
adaptive; per-OS specifics stay in the platform refs Setup loads
regardless.
- SKILL.src.md: Commands table lists the variants; Setup step 2 reads
the variant instead of the web file when the platform is native.
- audit.md / adapt.md: Platform sections replaced with a one-line
web-only guard pointing at the variant.
- animate.md / layout.md: Platform sections deleted; the Motion and
Layout sections of the already-loaded platform refs carry that
content. Web users now pay zero tokens for the platform axis in
these files.
- Skill-behavior scenario 15 pins the route-instead behavior (passes
live on claude-sonnet-4-6); CLAUDE.md documents the variant
convention.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Phrase command-reference routing as one rule, not rule-plus-exception
Copilot review catch: step 2 said "MUST read reference/<command>.md"
and then carved out the native variant, which invites loading both
files. Now a single rule: read the web reference or the table's native
variant, one file, not both. Scenario 15 re-verified live. Applied with
AI assistance (Claude Code) under maintainer direction.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Anchor native runs in animate/layout, drop loaded-refs assumption
Review-thread fixes, applied with AI assistance (Claude Code) under
maintainer direction:
- Greptile: deleting the animate/layout Platform sections left native
runs alone with web tooling instructions (CSS keyframes, GSAP, Grid,
clamp()). Restore a one-line anchor in each pointing at the loaded
platform reference's Motion / Layout section (~20 tokens, not the old
restatements).
- Bugbot: audit.native.md and adapt.native.md asserted the platform
refs were "already loaded in Setup", but the command reference loads
at step 2, before step 5. Now they instruct: read the platform
reference first if Setup hasn't already.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Carry the native-variant rule into routing rules 2 and 3
Bugbot catch: Setup step 2 routed native projects to the variant, but
routing rules 2 and 3 (the operative text at command time) still said
to load the generic reference file. Both now reference the same
one-file variant rule. Applied with AI assistance (Claude Code) under
maintainer direction.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Point animate/layout native anchors at the files, not "loaded" refs
Bugbot catch, same class as the variant wording fix: the anchor lines
said "the loaded platform reference" but command files load at step 2,
before the platform refs at step 5. Both anchors now name the files and
instruct reading them first if Setup hasn't already. Applied with AI
assistance (Claude Code) under maintainer direction.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* Add a platform axis (web / ios / android / adaptive) to the skill
Orthogonal to register: register decides whether design IS or SERVES the
product; platform decides the delivery target and which native conventions
apply. Set `## Platform` in PRODUCT.md; a missing field defaults to `web`,
so legacy projects are unaffected.
- extractPlatform() in skill/scripts/context.mjs (mirrors extractRegister);
the CLI appends a NEXT STEP directive to read the native reference(s).
`adaptive` (Flutter / RN / KMP shipping both iOS and Android) loads both
ios.md and android.md.
- New reference/ios.md (Apple HIG distilled) and reference/android.md
(Material 3 distilled); reference/web.md is a thin pointer. The native
refs frame register's role as narrow: platform conformance is the bar,
brand lives in the expressive layer the platform gives you, never by
breaking the rails.
- Setup step 5 loads the native reference(s) when platform is native. Live
mode and the detect CLI stay web-only, gated off ios/android/adaptive.
- init asks platform right after register; adapt/audit/animate/layout carry
short platform divergence notes; all secondary spots thread `adaptive`.
- a11y stays in audit.md (loading it at design time makes output timid), so
the native refs carry no Accessibility section; audit.md's Platform
section owns native a11y.
- Tests: extractPlatform unit coverage + skill-behavior scenario 10
(PRODUCT.md platform ios -> agent loads ios.md).
Source-first: only skill/, scripts/, tests/, CLAUDE.md, NOTICE.md, the
changelog and version are committed; the sync workflow regenerates the
provider trees and ./plugin on merge.
ios.md / android.md are distilled from the MIT-licensed
ehmo/platform-design-skills; attribution in NOTICE.md.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Address review: gate web tools on native platforms, drop version churn
Maintainer-review fixes applied with AI assistance (Claude Code), on top
of the rebased platform-axis commit:
- Design hook (post-edit and Cursor pre-edit) now resolves the project
platform via loadContext + extractPlatform and skips its web rule scan
for ios / android / adaptive projects, so React Native / Flutter code
never draws web-shaped findings (new hook-lib resolveProjectPlatform /
isNativePlatform helpers, covered by unit and subprocess tests).
- context.mjs CLI warns on an unrecognized ## Platform value (e.g. a
toolchain name like `flutter`) instead of silently defaulting to web;
extractRegister / extractPlatform now share extractSectionValue.
- Removed reference/web.md: nothing loaded it; CLAUDE.md carries the
"web has no extra rulebook" explanation.
- init.md: skip live-mode config (Step 6) for native platforms; note the
per-app PRODUCT.md pattern for repos shipping web + native.
- android.md: Material-everywhere apps that also ship on iPhone still
owe iOS OS guarantees (safe areas, Reduce Motion, edge-swipe back).
- ios.md: reworded a design-time line that framed Dynamic Type as an
accessibility check (a11y stays owned by audit.md).
- Renumbered the new skill-behavior scenario to 14 after main's 10-13;
updated CLAUDE.md scenario list; added android + unrecognized-value
CLI test cases.
- No version or changelog changes: versioning happens at release time.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Tighten platform reference prose
Editorial pass on the platform-axis text, applied with AI assistance
(Claude Code) under maintainer direction:
- ios.md / android.md rewritten to house style: single-line paragraphs
(no hard wraps), one-sentence scope intro, deduplicated intro/slop-test,
register-compression down to two sentences. In-file attribution
paragraphs removed (NOTICE.md owns attribution); "read on top of the
register reference" cruft removed (SKILL step 5 and the context.mjs
directive already say it). Bans sections dropped: they restated the
rules above them; the two additive items (tab-bar overload,
hover-dependent affordances) folded into rules. ~40% smaller each.
- Sub-command Platform sections (adapt, audit, animate, layout), SKILL
step 5, init.md platform prose, and the context.mjs directive trimmed
the same way.
Build (prose validators, counts) and both test runners green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Treat an empty PRODUCT.md section as absent, not the next heading
Copilot review catch: extractSectionValue read the next `## ...` heading
as the section value when a field was left empty, which made the CLI
warn "value `## Product Purpose` is not recognized". Stop at the next
heading and return null instead. Regression tests for extractPlatform,
extractRegister, and the CLI warning path. Applied with AI assistance
(Claude Code) under maintainer direction.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Only read a token list of both native targets as adaptive
Bugbot catch: after the exact platform tokens failed, any Platform line
containing the words ios and android was classified adaptive, so
negated or explanatory prose ("web only, not ios or android") silently
loaded both native refs and skipped the hook, with no warning. The
combo parse now accepts only list separators and the two platform
words; anything else falls through to the CLI's unrecognized-value
WARNING. Regression tests added. Applied with AI assistance (Claude
Code) under maintainer direction.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Paul Bakaus <paul.bakaus@gmail.com>
* Don't force init on scoped commands when PRODUCT.md is missing
Setup step 1 told the agent: "If it reports NO_PRODUCT_MD, stop and
follow reference/init.md before doing anything else." For a project with
no PRODUCT.md, that turned every scoped request (polish, critique, audit,
layout, ...) into a full from-scratch init detour. The user asks to
polish one button and the skill instead starts writing PRODUCT.md from
the beginning. Faced with that gate, agents also frequently abandon the
command and do an ad-hoc pass without loading the command reference.
Make the gate command-aware. A missing PRODUCT.md still routes into init
for the from-scratch build flows where captured product context is the
point (init, craft, shape). For any other command, a scoped request
against existing code, the code is the context: proceed with the
requested command, infer the register from the surface in focus, and
offer /impeccable init once as a suggestion rather than a blocker.
- skill/SKILL.src.md: rewrite the step 1 NO_PRODUCT_MD rule; reconcile
the no-argument routing rule so it leads the menu with init instead of
silently jumping into it; extend the craft init-then-resume footnote to
cover shape, now also a from-scratch flow.
- skill/scripts/context.mjs: soften the NO_PRODUCT_MD message to defer to
the step 1 rule instead of "Stop the current task"; refresh the stale
file-level JSDoc that still described the old empty-stdout signal.
- tests/skill-behavior/scenarios.test.mjs: add scenario 10 (scoped
command + no PRODUCT.md proceeds without forcing init) and scenario 11
(shape + no PRODUCT.md still diverts into init). Scenario 1 (craft
diverts) stays green and pins the build path.
Source-only per repo convention; provider and plugin copies are
regenerated by the maintainer's build:skills sync.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Fix missing-context routing for build intent
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Paul Bakaus <paul.bakaus@gmail.com>
* Fix live variant cycling hydration mismatch on SSR frameworks
Drive variant visibility and range/toggle --p-* custom properties through
an injected session stylesheet instead of mutating hidden/style on
server-rendered variant divs. Fixes flaky nextjs-app-router expectConsoleClean
failures (issue #287), same pattern as scroll-anchor (#276) and pick-cursor (#286).
Co-authored-by: Cursor <cursoragent@cursor.com>
* Refactor variant-state stylesheet for readability
Extract named display constants (VARIANT_HIDE_DECL / VARIANT_SHOW_DECL) and
small variantStateSelector / variantParamDecls helpers so the rule-building is
self-documenting. Restore the scroll-lock comment to startScrollLock. No
behavior change; regression guards updated to match.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix: keep variant-state stylesheet in sync on first-reveal and paramless cycle
Stop refreshParamsPanel from removing the injected variant-state sheet
during GENERATING first-reveal, and re-sync the sheet when cycling to a
paramless variant so stale --p-* rules do not persist. Harden the
updateVariantStateStylesheet guard to num == null || num < 1.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix: apply tuned --p-* inline for client-mounted Svelte component variants
Svelte component sessions mount into [data-impeccable-component-mount]
with no [data-impeccable-variant="N"] wrapper for the state stylesheet to
target. Restore inline --p-* on the client-mounted element for range/toggle
params while keeping the SSR div path on the injected stylesheet.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
A file-scoped wildcard ignore (add-value <rule> "*" --file <glob>) silently no-op'd for rules with no extractable value, such as side-tab. isIgnoredFindingValue bailed on an empty value before the wildcard/file-scope branch could run.
Require a value only on the specific-value path; let the scoped wildcard match on rule + file. Mirrored in skill/scripts/hook-lib.mjs for CLI/hook parity.
* Fix Pi global install path
* Simplify Pi skills-path helpers and consolidate tests
One userProviderSkillsDir helper owns the HOME_SKILLS_DIR_OVERRIDES
lookup, read paths share existingSkillsDirs, and the five Pi install
tests collapse into two that keep the same coverage: global detection
plus the agent-path write, and project scope in a home-rooted repo.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Respect requested scope when resolving Pi skills dirs
An explicit install scope now narrows providerSkillsDirCandidates to
the matching layout, so a project-scope install in a home-rooted repo
no longer matches an existing global Pi install and get swallowed by
the already-installed refresh path. Update/check flows still probe
both layouts since they have no scope. Covers the T-Rex repro in the
home-rooted regression test.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Refresh every existing Pi layout on unscoped update
deduplicateProviders keeps one entry per existing layout instead of
only the first, so unscoped check/update refresh both ~/.pi/agent/skills
and ~/.pi/skills when a home-rooted repo holds copies in each. Home-dir
detection now compares realpaths, since findProjectRoot resolves
symlinks while homedir() does not.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Abdul Wahab <abdulwahab@Abduls-MacBook-Pro-2.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
The PostToolUse hook was writing hook.cache.json after every edit, even
when nothing was scanned or recorded. Gate the persist to earned writes
only, and key the cache to the edited file's project root when the
session starts from an umbrella directory.
Fixes#344, #305
Co-authored-by: Abdul Wahab <abdulwahab@Abduls-MacBook-Pro-2.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
Codex loads bundled plugin lifecycle hooks from `hooks/hooks.json` using a
strict schema that accepts only the top-level `hooks` field. The
plugin-packaged manifest carried a top-level `description`, so Codex rejected
the whole manifest with `unknown field description, expected hooks` and the
post-edit design detector never registered (issue #330).
Drop `description` from `buildClaudePluginHooksManifest()` and regenerate
`plugin/hooks/hooks.json`. The Claude Code plugin path is unaffected (it only
reads the `hooks` object). Add a regression assertion for the plugin artifact
and bump the skill version to 3.9.1.
Claude-Session: https://claude.ai/code/session_013GTTHY6uHwESUyAgjUEm7x
Co-authored-by: Claude <noreply@anthropic.com>
* Add codex-grid-background detector rule
Detects the Codex two-axis grid-line background tell: a single background
value carrying two or more hairline `linear-gradient(... 1px, transparent
1px)` layers (one per axis), usually paired with a repeating
`background-size` cell. Gated behind --gpt like the sibling codex tells,
off by default.
Counts hairline stops within a single background declaration (not across
the page) so unrelated single-axis ruled lines don't add up to a false
flag, and matches the stop directly rather than parsing whole gradient
layers, since colors like oklch(...) carry nested parens.
Extends the gpt-tells fixture with one flag case and two pass cases
(single-axis rule, two-color blend), regenerates the browser detector
bundle, and bumps the rule count 44 -> 45.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Require tiling background-size for codex-grid-background
Address review: two hairline gradients alone draw a fixed crosshair, not a
grid. Scope detection to a single style block (CSS rule body or inline
style attr) and require both >=2 hairline stops AND a tiling
`background-size` px cell in the same block, matching the skill rule's
"plus background-size" wording. Add a crosshair-without-tiling pass case.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Scope codex-grid-background hairline count to background values
Address review: count hairline stops only inside background/background-image
declaration values, not the whole style block, so a hairline in an unrelated
property (mask-image, border-image) can't stand in for the grid's second
axis. Add a bg+mask-image hairline pass case.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Fix: preserve external skills symlink on first install (#295)
* Fix review comments: target-based in-project link detection (#295, #308)
- isInProjectProviderLink now inspects the symlink TARGET lexically instead of comparing shared realpaths, so two providers pointing at the same external dir are no longer misflagged as in-project (cursor High / greptile P1).
- A dangling in-project cross-provider link is now correctly replaced with a real per-provider dir (cursor Medium).
- Adds regression tests for both scenarios.
* Fix React hydration mismatch from live pick-cursor class on SSR roots
Entering pick mode toggled a `impeccable-live-pick-cursor` class on
`document.documentElement` (and the insert-axis cursor wrote an inline
`style.cursor` on it). `<html>`/`<body>` are server-rendered by frameworks
like Next.js App Router, so a client-only attribute the server HTML never
emitted makes React 19 log "a tree hydrated but some attributes of the server
rendered HTML didn't match" on the next Fast-Refresh re-render. It surfaced as
a console.error that flaked the nextjs-app-router live-e2e fixture's
expectConsoleClean probe.
This is the same root-cause class as the scroll-anchor lock fixed in #276
(client mutation of a hydrated SSR root), but a separate offender that fix did
not cover. Apply the same shape: drive the pick / insert cursor entirely
through the textContent of one injected `<style>` keyed by PICK_CURSOR_STYLE_ID,
never by a class or inline style on `<html>`. Same computed effect (global
`cursor` rule, reverted inside the overlay chrome), recreated on activation and
removed on teardown.
Regression guard updated to pin the new shape: no
`document.documentElement.classList.*` mutation anywhere in the overlay, the
cursor applied through the injected style, and the style removed by id on exit.
Verified end-to-end: the nextjs-app-router live-e2e fixture now passes the full
click -> Go -> cycle -> accept cycle with a clean console.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Remove now-dead pageInteractionCursorActive flag
The flag's only reader was the old inline-style cleanup branch in
syncPageInteractionCursor, which the stylesheet refactor removed. It is now
write-only, so drop the declaration and both writes (Greptile review). No
behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Add inline, in-file ignore comments for the detector (issue #283)
Complement config ignores with eslint-disable-style waivers that live where
they apply and travel with the file when it leaves the repo. The motivating
case is a generated/exported standalone document that legitimately uses a
first-party brand typeface (on the overused-font list) and is later scanned
without .impeccable/config.json present.
Marker is comment-syntax-agnostic (works in //, /* */, <!-- -->, #, {/* */}):
impeccable-disable <rule>[, <rule>...] [-- reason | : reason] whole file
impeccable-disable-line <rule>... same line
impeccable-disable-next-line <rule>... next line
Bare directive or * means every rule; reason is optional and discarded at
scan time. Behavior is suppression, for parity with config ignores.
Implementation:
- New pure module cli/engine/shared/inline-ignores.mjs (parser + filter, no
Node deps). Static-HTML findings have no line number, so only whole-file
directives apply there -- exactly the standalone-document case; the
regex/text engine additionally honors the line-scoped forms.
- Wired into detectText and detectHtml, gated by options.inlineIgnores.
- detect CLI applies inline ignores by default; --no-inline-ignores skips
just them, --no-config skips config and inline ignores together.
Docs: config.md (new section), detector.md, README. skill/reference/hooks.md
reversed its prior "inline comments are not supported" guidance and now points
the agent to inline waivers for the travels-with-the-file case. Changelog 3.x.
Tests: tests/inline-ignores.test.mjs (parser units, detectText/detectHtml
integration, CLI end-to-end), registered in the detector suite.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Reconcile design hook wording with inline ignores
Two hook-side fixes prompted by review of the new inline-ignore feature:
1. Clean-ack steer line. The old line ("Keep typography hierarchy, spacing
rhythm, and color contrast intentional on the next change.") read as an
odd non-sequitur after "No anti-patterns." Reworded the whole clean ack to
say what it means: a clean scan only clears the deterministic rule set, not
overall design quality, so keep following the design system and skill
guidance. Now: "Design hook scanned X. No deterministic design-quality
issues found. That does not mean the design is good: keep following the
project design system and the impeccable skill guidance."
2. Directive footer. It still told the agent "Do not add source comments such
as `impeccable: ignore`; those pollute the code and do not suppress hook
findings." That is now misleading: the hook runs the same detector engine
as the CLI, which honors inline `impeccable-disable` waivers, so they DO
suppress hook findings (consistent with config ignores, which filterFindings
already honors). Reworded to: don't silence a real finding to skip fixing
it; suppress only after the user confirms intent; prefer a config ignore,
and reach for an inline `impeccable-disable <rule>` comment only when the
waiver must travel with a file that leaves the repo.
Added a hook test asserting an inline `impeccable-disable-line` comment makes
the hook scan the file clean (locks in the cross-cutting behavior), and updated
the clean-ack / footer assertions to the new wording.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Address review on inline-ignores parser
- Case-insensitive fast-path bail-out (Cursor): the cheap substring guard was
lowercase-only while DIRECTIVE_RE has the `i` flag, so a mixed-case marker
like `Impeccable-Disable` skipped parsing entirely and never suppressed.
Switched the guard to `/impeccable-disable/i.test(...)`. Added a regression
test.
- Removed the unreachable `-->` branch from TRAILING_CLOSER_RE (Greptile):
`--+>` already matches `-->` and any longer dash run.
- Replaced the always-truthy lazy-match + `if (sep)` reason strip with an
explicit first-separator slice (Greptile): clearer and drops the dead branch.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Align inline-ignore line numbering with the detector (CRLF/CR endings)
parseInlineIgnores split lines with /\r\n|\r|\n/, but detectText numbers lines
with split('\n'). On classic `\r`-only endings the two diverged, so a
disable-line / disable-next-line directive could key a different line than the
finding it should waive (Cursor review). Split on '\n' only, matching the
detector exactly; the directive regex already excludes '\r', so a trailing '\r'
on CRLF files is never captured into the rule list. Added a CRLF regression test
through the real detectText.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two Cursor Bugbot Medium findings on the merged monorepo context PR:
- Excluded packages still listed: discoverTargetCandidates added every glob
match but never applied negated workspace patterns, so an excluded package
(e.g. "!packages/internal") showed up as a selectable target even though
resolveWorkspaceProjectRoot sends it back to the repo root. Now filtered
with the same isExcludedByWorkspacePattern check the resolver uses.
- Empty app list blocks root: resolveTargetSelection returned
TARGET_SELECTION_REQUIRED whenever projectRoot === repoRoot, even with zero
discoverable child apps (e.g. `workspaces: ["."]`), leaving an unanswerable
prompt. It now returns null (use the repo root as the project) when there
are no candidates.
Also documents two Greptile P2 clarity notes (the four contextSourceStatus
labels incl. the dual meaning of 'fallback', and the deliberate
isMonorepoRoot-before-hasGitBoundary ordering in findMonorepoRoot).
Adds regression tests for both behaviors.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Context files (PRODUCT.md / DESIGN.md) resolve child-first then fall back to the repo root, and /impeccable live lets the user pick a child app in a monorepo. Single-app behavior is unchanged. Closes#202. Co-Authored-By: abdulwahabone
* Add GitHub Copilot hook support (CLI + cloud agent)
Wire the Impeccable design detector into GitHub Copilot's hook system so
direct file edits get the same post-edit design feedback the Claude Code,
Codex, and Cursor harnesses already receive.
GitHub Copilot's contract differs from the existing harnesses (verified
against Copilot CLI 1.0.63):
- Repo-level manifest at `.github/hooks/impeccable.json` (read by both the
CLI, once committed to the default branch, and the cloud/app agent).
- Flat `postToolUse` entries with `bash`/`timeoutSec` and a full-match
`matcher` regex; the file-editing tools are `edit` and `create`.
- The stdin event uses camelCase `toolName`/`toolArgs`, where `toolArgs` is
a JSON *string* carrying the touched file under `path`.
- Context is injected via a top-level `additionalContext` string.
Changes:
- hooks.js: buildGitHubHooksManifest() + route `github` in hooksJsonFor().
- providers.js: emitHooks/hooksManifestRel for the github provider.
- hook-lib.mjs: detect the github harness, normalize the camelCase event
(parse the JSON-string toolArgs -> tool_input.file_path), and emit the
`additionalContext` payload shape.
- hook-admin.mjs / skills.mjs: install + idempotent-repair the
`.github/hooks/impeccable.json` manifest (bash-aware marker stripping).
- hooks.md: document GitHub Copilot as a supported harness.
- Tests for the builder, routing, event normalization, and end-to-end run.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Cover Copilot apply_patch edits in the hook (live-verified)
The first cut matched only `edit|create`, the tool names `copilot -p` uses.
A live trace against Copilot CLI 1.0.63 in an interactive session showed it
edits files via `apply_patch`, whose toolArgs is a raw OpenAI-format patch
string (`*** Begin Patch` / `*** Add File:`), not JSON. With the narrow
matcher the hook command never ran.
- hooks.js / hook-admin.mjs: matcher -> `edit|create|apply_patch`.
- hook-lib.mjs: normalizeGitHubEvent now routes apply_patch's raw patch
string into tool_input.command (reusing the existing parseApplyPatchPaths /
resolveTargetFiles plumbing) and only JSON-parses toolArgs for the
edit/create/view tools. tool_name is normalized to apply_patch so the patch
path is extracted even if a future build relabels the tool.
- Tests: apply_patch matcher assertions, event normalization, and an
end-to-end runHook covering the interactive/cloud path.
Verified live: a trusted interactive `apply_patch` edit fires the hook and
returns the expected `additionalContext` design reminder.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Address review feedback + add changelog entry
- hook-lib.mjs (Bugbot, low): looksLikeApplyPatch no longer misroutes an
edit/create event whose edited *content* contains apply_patch markers. A
real apply_patch payload is a raw string that does not parse as JSON; an
edit payload is a JSON object, so only non-JSON-object strings are treated
as apply_patch. Edit events keep extracting `path`. Adds a regression test.
- skills.mjs (Bugbot, medium): document why `.github` is intentionally
excluded from hookScriptPathForProvider. Its hook manifest is committed and
shared (read by the Copilot cloud agent and teammates), so the command must
stay portable via `$(git rev-parse ...)`; rewriting it to a machine-local
absolute path would break those. GitHub skills are project-scoped, so the
project-relative path resolves.
- changelog: add an Upcoming (v3.x placeholder) entry for the Copilot hook.
Version is not bumped yet (batching with other changes).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Guard plugin/skill version drift in the build (issue #274)
The Claude Code marketplace installs from the committed ./plugin subtree,
so a version disagreement between the hand-edited manifests and the
generated subtree ships stale content under a wrong version. This is the
class of bug reported in #274: a version bump that doesn't regenerate
./plugin (e.g. PR #252, where root plugin.json was 3.7.0 while
plugin/.claude-plugin/plugin.json was still 3.6.0) merges a drift window
onto main, and marketplace/Cowork installs pull the stale subtree.
Add a build-time validator that treats root .claude-plugin/plugin.json
as the source of truth and fails the build if any of these disagree:
- .claude-plugin/marketplace.json plugins[0].version (hand-edited; the
post-merge sync workflow never bumps versions, so it can't repair a
mismatch here)
- plugin/.claude-plugin/plugin.json version (generated subtree)
- plugin/skills/impeccable/SKILL.md frontmatter version (bundled skill)
It only fires on an inconsistent bump; PRs that don't touch versions keep
every file in agreement and stay silent. The pure comparison lives in
scripts/lib/validate-plugin-versions.js with direct unit coverage; build.js
owns the logging and the non-zero exit. Documents the regenerate-on-bump
step in CLAUDE.md's Versioning section.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Harden version-drift collector against malformed/incomplete manifests
Address Greptile review on #278:
- Wrap every file read/parse in a sentinel helper (extractFromFile) so a
half-edited manifest — the exact state during a version bump — yields a
clean "could not parse (...)" diagnostic naming the file instead of a raw
JSON.parse stack trace out of build().
- Report a present-but-malformed root plugin.json, or one missing its
`version` field, as an explicit error. Previously `undefined` version
short-circuited the build wrapper's `source == null` guard and passed
silently. collectPluginVersions now returns an `errors` array; build.js
fails on errors + mismatches combined, and only the genuinely-absent root
manifest is a no-op skip.
Adds 4 unit tests: malformed checked manifest, malformed root, missing
version field, and the absent-root no-errors case.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Make SKILL.md frontmatter version read CRLF-tolerant
Address Cursor Bugbot review on #278: readSkillFrontmatterVersion only
matched `\n` delimiters, while the shared parseFrontmatter in
scripts/lib/utils.js accepts `\r?\n`. A bundled SKILL.md saved with CRLF
line endings would parse to a null version and trip a false mismatch
against root plugin.json even when the version line is correct.
Match the shared parser's `\r?\n` tolerance and drop the `$` anchor on
the version line (it would not match before a `\r`). Adds CRLF coverage
for both readSkillFrontmatterVersion and collectPluginVersions.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Re-trigger CI (no file change)
CI did not fire for 5cda9f6b; force a fresh run on the current tree.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Fix React hydration mismatch from live scroll-lock on SSR roots
The live overlay's startScrollLock disabled the browser's scroll
anchoring by setting `overflow-anchor: none` as an inline style on
`<html>` and `<body>`. On frameworks that server-render those roots
(notably Next.js App Router), that client-only inline style desyncs from
the server HTML, so React 19 logs "a tree hydrated but some attributes
of the server rendered HTML didn't match" on the next Fast-Refresh
re-render. It surfaced as a flaky failure of the nextjs-app-router
live-e2e fixture's expectConsoleClean probe.
Inject the suppression as a `<style>` rule keyed by a stable id instead
of mutating inline styles on hydrated host elements. Same computed
effect, but React no longer sees a client-only attribute on `<html>` /
`<body>`. The rule is recreated on every startScrollLock and removed on
teardown, so reload survival (driven by the persisted scroll key) is
unchanged.
Adds a regression guard pinning the new shape (no inline overflowAnchor
mutation on html/body; injected <style> created and removed by id).
Verified end-to-end: the nextjs-app-router live-e2e fixture now passes
the expectConsoleClean probe deterministically.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Relax regression-guard regex spans to {0,400}
Address Greptile review: the {0,200}/{0,220}/{0,160} character-span
limits between the injected-style constructs were tight enough that an
innocent refactor or added comment inside startScrollLock could silently
break the shape-check. Widen each segment to {0,400}; the guard still
passes on the fix and still fails when the inline html/body overflowAnchor
mutation is reintroduced.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The bundled detector's cli/main.mjs imports ../../lib/impeccable-config.mjs,
which in the source CLI resolves to cli/lib/impeccable-config.mjs. The skill
build only copies cli/engine/** into scripts/detector/**, leaving that
dependency behind, so from the bundled scripts/detector/cli/main.mjs the same
import resolved to scripts/lib/impeccable-config.mjs and failed with
"Cannot find module .../lib/impeccable-config.mjs". /impeccable critique (and
any detector-backed command) crashed on startup for every provider since #252.
Teach the detector bundler to copy out-of-bundle engine dependencies into the
skill's scripts/lib/, and add a build test that walks every bundled script and
asserts each relative import resolves to another bundled file, so a future
out-of-bundle dependency fails the test instead of the user.
Skill v3.7.1 (patch). CLI unchanged — the engine resolves fine in the CLI's
own tree; only the skill bundling was wrong.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
On Node v24.16.0 / v26.1.0+, `impeccable install` printed "Downloading
impeccable skills...", exited 0, and installed nothing. A Node streams
regression (nodejs/node#63487) made pause()/resume() no-ops on destroyed
streams, stalling extract-zip's yauzl/fd-slicer read stack partway through;
its promise never settled and the process exited clean with no error.
Swap extract-zip for fflate across both extraction call sites
(downloadAndExtractBundle, copyOrExtractLocalBundle) via a new extractZip
helper. fflate decompresses from an in-memory buffer and never touches the
fs stream path, so it is immune on every Node version. It is pure JS with
zero dependencies, so the Windows fix from #198 (no `unzip` binary) holds.
Unlike extract-zip, fflate is actively maintained.
Because extractZip writes entries itself, it guards against zip-slip (`../`
entries escaping the target dir). Tests add a many-file regression guard
(fails on partial extraction) and a zip-slip rejection test.
Verified end-to-end: the real 1,194-file universal bundle extracts and
installs completely.
Fixes#250.
Co-authored-by: Claude <noreply@anthropic.com>
Paul Bakaus
Dustin PersekandGitHub
