Whole-plan coherence review (Plan 3a Layout + Plan 3b Motion) found that
Layout's keyframe gutter and Motion's Timing row independently derived an
element's start/duration and disagreed whenever an element had animations
but no explicit data-duration: Motion correctly inferred the range from the
element's GSAP tweens, while Layout's keyframe gutter fell back to a naive
`duration ?? 1`, so clicking a keyframe percentage in Layout could seek to a
different absolute time than what Motion's Timing row displayed.
Extract deriveElementTiming (propertyPanelFlatTimingDerivation.ts) as the
single shared basis both paths now consume: FlatTimingRow (Motion) and
PropertyPanelFlat's own elStart/elDuration (Layout's keyframe gutter and 3D
Transform block). PropertyPanelFlat now recomputes this basis itself from
its own element/gsapAnimations props instead of trusting the parent's naive
value, so PropertyPanel.tsx (and its legacy non-flat panel) is untouched.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Thread the Layout-group values through PropertyPanel -> PropertyPanelFlat
and add the third FlatGroup to the one-open/pin accordion (unconditional,
matching legacy Layout). Default-open Layout when neither Text nor Style
applies.
Fix the Flex double-render: the legacy StyleSections still renders its own
Flex Section, and the new flat Layout group renders its own LayoutFlexBlock.
Add an additive optional hideFlex prop to StyleSections and pass it on the
flat path so Flex renders exactly once (from the flat Layout group). Non-flat
callers omit it and are unchanged.
Extract the shared onLivePreviewProps closure into gsapLivePreview.ts (it was
duplicated inline in the legacy path) so PropertyPanel.tsx stays within the
600-LOC studio gate.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The flat inspector's Text FlatGroup rendered unconditionally, showing an
empty "Text" header for non-text elements (image, video, etc). Gate it on
isTextEditableSelection(element) so it disappears entirely when there's no
text to edit.
Also, the legacy multi-field TextSection (used as a fallback when an
element has 2+ text fields) rendered its own internal "Text" heading
nested inside the new flat Text FlatGroup, producing a doubled "Text"
heading. Add a hideOwnHeading prop to TextSection (default false, so its
other — legacy, non-flat — call site is unaffected) and pass it from
FlatTextSection's fallback path.
Documents the shared-pattern context (3rd copy of "resolve relative
data-start", after runtime startResolver.ts and the SDK's own
getElementTimings) and explains when the raw parseFloat fallback in
resolveStart's else branch can actually fire (a malformed grammar string
with a leading number). Adds a test pinning the "reference target exists
but its own timing is unresolvable" branch, which existing tests didn't
cover (only "target doesn't exist" was tested).
Cross-checked the negative-offset clamp concern raised in review: the
SDK's own resolveReferenceStart (session.ts) also clamps to
Math.max(0, ...), so this stays consistent with its sibling — no code
change needed there.
Same bug class as the SDK's getElementTimings fix (#2092): data-start can be a
relative-reference expression ("intro", "intro + 2"), not just an absolute
number. The old code did a raw parseFloat on it, so any reference silently
resolved to undefined instead of an actual time.
Also: this function never read data-duration at all (only data-start/data-end
literally), so a reference to a duration-authored (not end-authored) clip was
unresolvable regardless of the parseFloat bug — resolving a reference needs
the target's END, which for a duration-authored clip requires start+duration.
Both fixed together via the shared parseStartExpression grammar parser
(@hyperframes/core/runtime/start-expression), with the same cycle-guard
pattern as the SDK fix. Reference resolution against other elements is scoped
to this file's existing findById (bare data-hf-id lookup).
6 new tests: duration-based end resolution, relative reference (with and
without offset), missing target, and a mutual-cycle termination check.
## What
`applyPositionEdits(doc)` in `@hyperframes/core/runtime/position-edits` guarded each candidate element with `instanceof HTMLElement`. `doc` is frequently an iframe's document (the SDK's edit preview, any host embedding a composition), whose elements are `HTMLElement` instances of *that frame's realm* — never this module's. The check silently no-ops on every single element cross-realm, so bulk position edits never apply inside an iframe.
## Why
Found during an audit of `@hyperframes/sdk`'s surface against pacific's movio integration. Pacific's `canvas-react` code has an explicit workaround comment for this exact bug: *"Upstream fix would be duck-typing in `@hyperframes/core` — until then, all host code must use this wrapper."* Every iframe-hosted consumer has had to reimplement the bulk-apply loop themselves to avoid it.
## How
Use the document's own realm's `HTMLElement` constructor (`doc.defaultView?.HTMLElement`) instead of the module-scope global. Duck-type on `.style` when `defaultView` is unavailable (a detached/synthetic document). The single-element `applyPositionEditToElement` was already realm-safe — only the bulk wrapper had the bug.
## Test plan
- [x] New regression test using a real jsdom iframe — confirmed it fails on the old `instanceof HTMLElement` check (0 applied, expected 1) and passes with the fix
- [x] Full existing `positionEdits.test.ts` suite passes (14/14)
- [x] Full `@hyperframes/core` suite passes (81 files / 1131 tests)
- [x] `bun run build` clean (core + full workspace, incl. studio)
A figma text node whose box is shorter than its line-height carries
vertically-trimmed (cap-to-baseline) bounds. The mapper positioned the box
at those bounds but let the browser lay glyphs with half-leading, pushing
them ~6px low on a 70px font (glyph-centroid measurement against figma's
own render: +9.1px vs figma's +3.4px inside the same pill). Emitting
text-box-trim: trim-both / text-box-edge: cap alphabetic reproduces the
trim in the render engine; post-fix centroid agrees within 0.4px and the
motion verifier's min window score improved 20.3 -> 25.3dB. Trim applies
only to single-line trimmed text; boxes matching their line-height are
untouched.
Skill: component imports now include a static fidelity self-check step
against figma's PNG export of the same node.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(cli): bump @puppeteer/browsers to ^3.0.6 to fix render hang on node >=24.16
`hyperframes render` (and `browser ensure --force`) hangs forever during
Chrome provisioning on Node >= 24.16 (repro'd on macOS arm64 / Node 26.5.0;
fine on Node 22). Root cause is a transitive extractor bug, not our logic:
@puppeteer/browsers@2.13.x install()
-> extract-zip@2.0.1 -> yauzl@2.10.0
A classic-stream backpressure regression (nodejs/node#63487, works 24.15,
breaks 24.16+) surfaces a latent fd-slicer destroy() bug in yauzl 2.x
(yauzl#169). The inflate read stream stalls partway through the first entry
large enough to cross the write highWaterMark (chrome-headless-shell's
1.86MB LICENSE.headless_shell, stalls at ~1.31MB), never emits `end`, so
stream.pipeline never settles and extraction busy-spins. The half-extracted
cache has no executable, so every later render re-enters
"Cached binary missing -> re-download" and hangs again (puppeteer#14957).
Fix: @puppeteer/browsers 3.0.2 dropped extract-zip/yauzl entirely (now uses
modern-tar). Verified 3.0.6 extracts chrome-headless-shell cleanly under
Node 26.5.0 and keeps the full API manager.ts uses (install,
getInstalledBrowsers, Cache, computeExecutablePath, detectBrowserPlatform,
Browser) with an identical on-disk cache layout. Cross-platform (the same
.zip/yauzl path affected Linux + Windows too).
Adds a regression guard asserting the pin stays on the extractor-free
major (>= 3) and never reintroduces extract-zip/yauzl.
Fixes#2103
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(cli): clarify extractor-guard wording — yauzl is an optional peer, not dropped entirely
Review note on #2104: @puppeteer/browsers 3.0.6 keeps yauzl as an optional
peer fallback (default extractor is modern-tar), so the regression-guard
comment + it-text shouldn't say it was 'dropped entirely'. Test assertions
(extract-zip + yauzl absent from `dependencies`) unchanged and correct.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A base `gsap.set(...)` written AFTER the tween calls is wiped on the next
soft reload: when a `from()` tween on the same target lazily initializes
during a backwards render (the studio rebind's progress(0.0001) kick), GSAP
reverts its internal isFromStart set, which removes the whole inline
`transform` — taking the base set's x/y with it. The from() tween then
re-parses the computed transform as identity and bakes x/y = 0 into the
GSAP cache, so every element previously moved in the studio snaps back to
its authored position whenever any other element is edited.
Emitting the global set BEFORE the timeline construction makes it part of
the pre-tween state the from() records, so every revert restores the moved
pose instead of stripping it.
- addAnimationToScript: global sets insert above the timeline declaration;
the new-id lookup now diffs content-based ids instead of assuming the
appended statement is last in source order.
- updateAnimationInScript: a legacy trailing global set is relocated above
the declaration whenever it's touched, healing files written before this
change on the next nudge.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
HF_DE_PARALLEL_STREAM was restored on every exit path, but the producer
server allows concurrent renders in one process — a router-eligible job's
mutation was still visible to an unrelated render already executing during
that window. Thread the router's decision as a per-render local instead of
a global env var; HF_DE_PARALLEL_STREAM stays as the manual opt-in.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- motionContextToDocs: escape regex metacharacters in arrayAfterKey /
scalarAfterKey key interpolation (safe today for \\w+ keys; now safe for
any future caller), and document balancedBlock's no-strings invariant.
- verify-motion.mjs: execSync shell string -> spawnSync with array args
(JSON.stringify is not shell escaping); verifier re-calibrated unchanged
(faithful render still PASS at min 20.30dB).
- command-failure-tracking: rebase folded the group-delegation skip into
upstream's recursive wrapCommand (HF#2033) — leaf commands now assert
their own flag tables, so `figma component --namee` is rejected at the
leaf while `--name` passes the group; heuristic invariant documented.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Two guarantees so figma-motion imports can't drift from the design again:
- motionContextToDocs(): raw get_motion_context response -> MotionDoc[],
in code. Parses the motion.dev snippets (the reliable encoding; the CSS
snippets stretch durations and can disagree), strips loop-wrap tail
keyframes (sub-ms segments at the window end are the loop reset, not
authored motion), preserves bezier eases verbatim. Fixture test uses the
verbatim response from a real Motion timeline whose translation was
frame-validated against Figma's own export_video render.
- skills/figma/scripts/verify-motion.mjs: mandatory post-render gate.
Compares motion-energy deltas between the render and the export_video
ground truth so static import fidelity cancels out and the score
isolates choreography. Calibrated on a faithful translation (min 20.3dB)
vs a diverging one (min 5.0dB); threshold 15dB.
The skill's Motion step now routes through both: no hand transcription,
no unverified completion.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Field lesson from translating a real Motion timeline: the two returned
encodings window durations differently, and keyframes at times ~0.9999
are loop-wrap resets, not authored motion. Hand-normalizing across
encodings and inventing visible returns produced a render that diverged
from Figma. The skill now mandates verbatim single-encoding translation,
wrap-via-repeat, and a frame-grid comparison against export_video ground
truth before completion.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
slugify("3D Object - Headphones") produced id="3d-object-headphones" —
valid HTML, but querySelector("#3d-…") throws (CSS idents cannot start
with a digit), which kills GSAP targeting and figma-motion translation
against imported components. uniqueSlug now prefixes digit-leading slugs
("n3d-object-headphones"). Found translating a real Figma Motion timeline.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Both found running the brand-loop guide end-to-end against the Simple
Design System:
- nodeToHtml subtracted the ROOT origin from every node's absolute bounds,
but CSS absolute positioning resolves against the nearest positioned
ancestor — every nesting level re-added its ancestors' offsets, drifting
nested content down-right and pushing deep children off-frame (hero
buttons invisible, pricing grid collapsed to one card). Children now
subtract their PARENT's box; regression test with a two-level tree.
- trackCommandFailures asserted unknown flags against the command group's
own (flagless) arg table even when the group was delegating to a
subcommand, so `figma component <ref> --name x` imported and THEN threw
"Unknown flag: --name". The assertion is now skipped when the first
positional names a subcommand; leaf and non-delegating behavior is
unchanged and covered by tests.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
