| name | optimize-ios-ui |
| description | Optimize iOS UI animations, haptics, and rendering performance. Dispatches animation-haptics-engineer (motion tuning + haptic placement) and performance-engineer (body evaluation, scroll, image handling, rendering) in parallel. Merges findings into a unified optimization report. Triggers on "optimize my animations", "improve haptics", "fix scroll jank", "optimize iOS UI performance", "make animations smoother", "add haptics", "tune springs". Also use proactively when a change touches animation, transition, gesture, haptic, or scroll code in an iOS project. |
Optimize iOS UI
Dispatch
This skill dispatches 2 specialist agents in parallel:
1. apple-ui-craft:animation-haptics-engineer
- Scope: all files in the optimization scope
- Reviews: spring parameters, animation timing, haptic coverage, Reduce Motion
- Focus: FEEL -- do animations communicate? Are haptics intentional?
2. apple-ui-craft:performance-engineer
- Scope: same files
- Reviews: body evaluation, lazy loading, image handling, rendering, scroll, animation frame rate
- Focus: SMOOTHNESS -- 60fps sustained, no hitches, efficient rendering
After both complete, merge findings:
- Cross-reference: if animation-haptics-engineer suggests a heavier spring that performance-engineer flags as causing hitches, note the conflict and suggest a compromise
- Order by impact (most noticeable improvement first)
- Group by screen/flow
Scope determination
| Arg | Meaning |
|---|
empty / diff | Uncommitted + staged changes (default) |
staged | Only staged files |
<file> or <directory> | Specific target |
all | Entire project (excluding generated code, Pods, build artifacts) |
When to use this vs. review-ios-ui
| Goal | Skill |
|---|
| "Does this look/feel Apple-native?" | review-ios-ui (visual design + a11y) |
| "Make animations smoother and add haptics" | optimize-ios-ui (motion + performance) |
| "The full treatment" | craft-ios-ui (all specialists) |
Output
Unified report with:
- Animation verdict (FLUID / ADEQUATE / STIFF / BROKEN)
- Haptic verdict (INTENTIONAL / SPARSE / ABSENT / NOISY)
- Performance verdict (SMOOTH / ADEQUATE / JANKY / BROKEN)
- The evidence mode, and whether Instruments actually ran. Every performance finding carries an
Impact: line with a metric and a delta, labelled measured or estimated -- an unmeasured performance claim sends someone optimising the wrong thing (references/review/02-evidence-pipeline.md)
- Findings in the canonical format (
references/review/01-finding-format.md); feel cannot be judged from a screenshot, so in Screenshots mode the animation and haptic verdicts are NOT ASSESSED
- Haptic coverage map (every interaction surface mapped to recommended haptic)
- Spring parameter recommendations (exact values)
- Performance fixes (concrete rewrites)
- Top 3 priority actions
Execution mode
Every agent this skill dispatches is pinned to Opus 5 (model: "opus") at dispatch -- the coding/review floor (owner directive 2026-07-24); the session conductor stays orchestrator-only. When the session model is already the strongest tier and the optimization scope is small, the orchestrator may run a specialist's review inline in the main context (foreground) instead of dispatching a separate agent, without weakening the read-only guarantee the reviewer agents carry.
Review ledger (write it, without asking)
At the end of every run, write .claude/apple-ui-craft/last-review.json in the
REVIEWED repo, and read any existing one first to produce a delta report (NEW /
RESOLVED / STILL OPEN / REGRESSED / IMPROVED). Schema, matching rules, and the
two fields that stop a narrow run from erasing a wide one (dimensions and
notAssessed): references/review/04-run-artifacts.md#review-ledger-schema-v1.
A missing or unreadable ledger is an empty prior run, never an error. A ledger
from a different scope is not a prior run for this scope.
Ultracode conductor mode
When the harness announces ultracode, this skill runs conductor-executor per references/_scaffolding/conductor-dispatch-protocol.md -- read that file before the first executor dispatch; it owns the dispatch mechanics, the fan-out doctrine (executor teams scale to natural breadth; the session-model agent caps do not apply to them), the executor prompt contract, and the validation gate. Without ultracode, run the standard 2-specialist dispatch above unchanged.
Split of labor
| Conductor (session model -- never delegated) | Conductor-selected executors (Sonnet 5 @ xhigh or Opus 5) |
|---|
| Animation/haptic/performance verdicts, spring-parameter judgment, conflict resolution (heavier spring vs hitch), apply/no-apply judgment, final report synthesis | Instrumentation sweeps: body-reevaluation candidates, animation inventory (curve/spring params per site), haptic-trigger inventory, scroll-container census; post-approval mechanical application of approved parameter changes |
Executor scoping (on top of the protocol's prompt contract)
- Reference set: absolute paths of
references/animation/*, references/interaction/*, references/haptics/*, references/performance/* + references/_scaffolding/version-floor-registry.md.
- Inline the severity scale and the 11-row a11y/perf gate from
agents/apple-ui-reviewer.md (rows 1-3 Reduce-Motion gating, row 5 compositor cost, rows 9-10 transition/flash safety -- the motion-relevant rows; sourced from references/accessibility/05-motion-accessibility.md, references/patterns/01-gotchas-anti-patterns.md, references/performance/01-swiftui-rendering.md).
- The
animation-haptics-engineer / performance-engineer specialists are pinned to Opus 5 at dispatch -- judgment reviewers, never grunt executors.
