AA hig-inputs
Design Apple-platform input interactions for touch, keyboard, pointer, Pencil, controllers, spatial gestures and supported device sensors, with accessible alternatives.
The skill teaches an AI agent to check for .claude/apple-design-context.md before asking questions and reuse existing context instead of requesting it again. Straightforward: don't ask twice.
The package contains 14 files with no critical issues. Scores are solid: quality 91, process 81. No broken links or lint errors. In practice, the skill works as an instruction to save tokens and time—the agent searches for context locally first, then acts. Wide platform support: Claude, Cursor, Gemini, DeepSeek, and eight others.
Worth installing if you work with the same project context repeatedly. If your context rarely changes and the file stays in place, you'll see the benefit. Otherwise it's just an extra step in the prompt.
Design Apple-platform input interactions for touch, keyboard, pointer, Pencil, controllers, spatial gestures and supported device sensors, with accessible…
As a process A 81/100 · Runs to the end — weak spots: progress reporting
The same skill appears in 1 more place: agentic-awesome-skills
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 14. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
edit-residuethe text marks something as outdated (lines 71): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 81/100
- 0Progress reporting. Says nothing while it works
- 50When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 70Inputs and preconditions. Inputs and preconditions are listed
- 85Steps. 29 steps, 1 vague phrases
- 100Tools and files. No external tools needed
- 100Result and completion. Output format and completion criterion are stated
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1842 tokens
- 100Running it twice. No mutating operations
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +1No license
- +2Single-language instructions
- +3Description length 168: enough signal without eating the budget
- +4Structure: 21 headings
- +3Step-by-step instructions: 29 items
- +3Output format is stated explicitly
- +4Has examples (0 code blocks)
- +4Reference files are cited in the instructions (13 of 13)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 91.