visionos-design-guidelines
by ehmoThe visionos-design-guidelines skill helps you apply Apple Vision Pro rules for spatial UI, eye and hand input, immersive spaces, windows, volumes, and accessibility. Use it when reviewing or designing visionOS interfaces that need comfort, correct placement, and platform-accurate guidance.
This skill scores 85/100, which means it is a solid directory listing candidate for users building or reviewing visionOS work. It gives agents enough trigger language, rule structure, and platform-specific guidance to reduce guesswork versus a generic prompt, though users should still expect a design-guidelines workflow rather than an executable tool.
- Clear activation scope for visionOS, Apple Vision Pro, RealityKit, spatial UI, and mixed reality tasks.
- Substantial, non-placeholder content with 50+ rules across categorized sections and severity levels, supporting concrete design decisions.
- Helpful operational framing in AGENTS.md and rules/_sections.md, including never-do constraints and quick-reference rule indexing.
- No install command or scripts are provided, so adoption is documentation-driven rather than tool-assisted.
- The repository is guidance-heavy and does not show end-to-end task workflows or examples for applying the rules in agent actions.
Overview of visionos-design-guidelines skill
What this skill is for
The visionos-design-guidelines skill helps you apply Apple Vision Pro design rules when you need to decide how a spatial UI should be laid out, how eye and hand input should work, or whether an immersive concept fits visionOS conventions. It is most useful when you are not just asking for inspiration, but need design constraints that affect comfort, reachability, and platform correctness.
Who should install it
Install visionos-design-guidelines if you are designing or reviewing a visionOS app, RealityKit experience, volume, window-based interface, or mixed reality prototype. It is especially useful for UI designers, product engineers, and AI agents that need to produce implementation-ready guidance instead of generic “make it immersive” advice.
Why it stands out
This skill is less about broad product theory and more about practical guardrails: where to place content, how to handle gaze-plus-pinch interaction, what not to head-lock, and how to avoid comfort or accessibility mistakes. The visionos-design-guidelines skill is strongest when you need a fast design check against critical spatial rules before you build.
How to Use visionos-design-guidelines skill
Install and load the right context
Use the visionos-design-guidelines install flow from your skills manager, then read the skill entry files first: SKILL.md, AGENTS.md, metadata.json, and rules/_sections.md. This skill does not depend on scripts, so the value is in the rule set and the category summaries.
Start with a concrete spatial brief
The visionos-design-guidelines usage works best when your prompt includes the interface type, user goal, and physical context. Good inputs sound like: “Design a visionOS onboarding flow for a productivity app using one window and one ornament,” or “Review this immersive dashboard for comfort, target size, and eye/hand interaction risks.” Weak inputs like “make this more Apple-like” usually produce generic output.
Turn a rough idea into a better prompt
For visionos-design-guidelines for UI Design, include:
- app surface: window, volume, immersive space, or mixed layout
- primary task: reading, selection, manipulation, navigation, or status display
- interaction model: gaze, pinch, drag, hand gesture, voice, or mixed input
- constraints: fixed window count, accessibility needs, comfort limits, or existing RealityKit code
A stronger prompt is: “Using visionOS guidelines, critique this two-panel dashboard for eye strain, field-of-view placement, and control sizes; then rewrite it as a spatially comfortable layout.”
Read files in this order
Use the visionos-design-guidelines guide as a repository-reading path:
rules/_sections.mdfor the severity-ranked rule mapSKILL.mdfor the detailed rules and anti-patternsAGENTS.mdfor purpose, scope, and usage cuesmetadata.jsonfor references and version context
If you are short on time, start with the critical categories: spatial layout, eye and hand input, and accessibility.
visionos-design-guidelines skill FAQ
Is this only for Apple Vision Pro?
Mostly yes. The visionos-design-guidelines skill is centered on visionOS and Apple Vision Pro conventions, though the rules are also useful for any spatial computing UI that uses gaze, hand input, or world-anchored layout.
How is it different from a normal prompt?
A normal prompt may give you a style answer. This skill gives you a rule-based design lens: distance, placement, target size, hover feedback, accessibility coverage, and immersion boundaries. That makes it better for review, implementation, and avoiding platform-specific mistakes.
Is it beginner-friendly?
Yes, if you can describe the screen or experience you want. You do not need deep visionOS knowledge to use visionos-design-guidelines, but you do need to provide enough context for the skill to judge spatial layout and interaction patterns accurately.
When should I not use it?
Do not rely on it for general mobile UI, web-only interfaces, or non-spatial product writing. It is also not the right fit if you need code generation without design review, or if your task has no visionOS, RealityKit, or mixed reality component.
How to Improve visionos-design-guidelines skill
Give it the real interface shape
The biggest quality jump comes from specifying what exists on screen and where it lives in space. Say whether you have a single window, multiple windows, a floating volume, or an immersive scene, and note what the user must do in the first 5 seconds.
Surface the constraints that change the answer
visionos-design-guidelines works better when you mention comfort, accessibility, and input constraints up front. Include things like minimum target sizes, whether the interface must support VoiceOver or pointer control, and whether the design needs to avoid frequent head movement.
Ask for a critique, then a rewrite
A useful workflow is: first ask for rule violations, then ask for a corrected version. Example: “Audit this visionOS onboarding against comfort, gaze feedback, and accessibility rules; then rewrite the layout and interaction sequence.” That usually produces more actionable output than asking for a generic design.
Use the first answer to narrow the next pass
If the initial result is too broad, refine by category: spatial layout, eye and hand input, windows, or accessibility. The visionos-design-guidelines skill improves when you iterate on one failure mode at a time, such as content placement, hover states, or world anchoring.