Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Desktop application accessibility expert -- platform APIs (UI Automation, MSAA/IAccessible2, NSAccessibility), accessible control patterns, screen reader Name/Role/Value/State, focus management, high contrast, and custom widget accessibility.
.claude/skills/community-access-desktop-accessibility-specialist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-07 | ✓→✗ | ▼ Worse | 64% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 25% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 10% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
Shared instructions
You are a desktop application accessibility specialist -- an expert in making desktop software fully usable by people with disabilities. You understand platform accessibility APIs, screen reader interaction models, and the complete lifecycle of accessible control design across Windows and macOS.
Knowledge domains: Python Development
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 16,520 | 17,946 | +9% | 1 | 1 | 0% | 2,558 | 3,197 | +25% | 0 | 0 | — |
case-01 | pass→pass | 14,240 | 13,480 | -5% | 1 | 1 | 0% | 2,488 | 2,746 | +10% | 0 | 0 | — |
case-02 | pass→pass | 22,533 | 14,206 | -37% | 1 | 1 | 0% | 2,997 | 2,927 | -2% | 0 | 0 | — |
case-03 | pass→pass | 12,478 | 14,850 | +19% | 1 | 1 | 0% | 2,524 | 3,235 | +28% | 0 | 0 | — |
case-04 | pass→pass | 13,478 | 14,433 | +7% | 1 | 1 | 0% | 2,228 | 3,067 | +38% | 0 | 0 | — |
case-05 | pass→pass | 13,532 | 15,385 | +14% | 1 | 1 | 0% | 2,308 | 3,031 | +31% | 0 | 0 | — |
case-06 | fail→pass | 13,276 | 8,977 | -32% | 1 | 1 | 0% | 2,338 | 1,999 | -14% | 0 | 0 | — |
case-07 | pass→fail | 11,339 | 14,895 | +31% | 1 | 1 | 0% | 1,680 | 2,753 | +64% | 0 | 0 | — |
case-08 | pass→pass | 14,985 | 17,838 | +19% | 1 | 1 | 0% | 2,868 | 3,528 | +23% | 0 | 0 | — |
case-09 | pass→pass | 13,716 | 16,526 | +20% | 1 | 1 | 0% | 2,006 | 2,751 | +37% | 0 | 0 | — |
case-10 | pass→pass | 15,617 | 18,290 | +17% | 1 | 1 | 0% | 2,593 | 3,250 | +25% | 0 | 0 | — |
case-20 | pass→pass | 9,548 | 9,037 | -5% | 1 | 1 | 0% | 1,695 | 2,168 | +28% | 0 | 0 | — |
case-11 | pass→pass | 7,783 | 9,153 | +18% | 1 | 1 | 0% | 1,202 | 1,828 | +52% | 0 | 0 | — |
case-12 | pass→pass | 6,959 | 7,752 | +11% | 1 | 1 | 0% | 1,061 | 1,567 | +48% | 0 | 0 | — |
case-13 | pass→pass | 10,603 | 9,064 | -15% | 1 | 1 | 0% | 1,732 | 1,781 | +3% | 0 | 0 | — |
case-14 | pass→pass | 3,930 | 4,312 | +10% | 1 | 1 | 0% | 620 | 1,052 | +70% | 0 | 0 | — |
case-15 | pass→pass | 9,291 | 16,687 | +80% | 1 | 1 | 0% | 1,695 | 3,275 | +93% | 0 | 0 | — |
case-16 | pass→pass | 6,264 | 4,479 | -28% | 1 | 1 | 0% | 1,005 | 1,073 | +7% | 0 | 0 | — |
case-17 | pass→pass | 10,689 | 11,845 | +11% | 1 | 1 | 0% | 1,721 | 2,365 | +37% | 0 | 0 | — |
case-18 | pass→pass | 16,047 | 16,367 | +2% | 1 | 1 | 0% | 2,738 | 2,926 | +7% | 0 | 0 | — |
case-19 | pass→pass | 12,044 | 13,555 | +13% | 1 | 1 | 0% | 2,037 | 2,512 | +23% | 0 | 0 | — |
case-22 | pass→pass | 13,227 | 14,605 | +10% | 1 | 1 | 0% | 2,173 | 2,566 | +18% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.