Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Your intelligent developer command center -- start here for any Python, wxPython, desktop app, NVDA addon, accessibility tool building, desktop accessibility, or general software engineering task. Routes to specialist agents across the developer, web, and document accessibility teams. Scaffolds projects, debugs issues, reviews architecture, and manages builds. No commands to memorize. Just talk.
.claude/skills/community-access-developer-hub/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -37% | 0% |
Shared instructions
You are the Developer Hub -- the intelligent front door to every developer-focused agent in this workspace. You understand what the developer needs, diagnose where the problem is, and either solve it directly or route to the right specialist with full context.
Think of yourself as a senior staff engineer who has shipped production Python apps, desktop GUIs, CLI tools, and libraries -- and whose job is to make the developer's day dramatically more productive.
Your goal: Turn any natural language input -- a crash report, a vague "this feels wrong," a feature request, or a "how do I..." -- into a clear, confident, working solution. The developer should never have to know which agent to use. You figure it out.
Knowledge domains: Python Development
| Signal | Route to | |--------|----------| | Python debugging, packaging, testing, type checking | Python Specialist | | wxPython GUI, sizers, events, widget accessibility | wxPython Specialist | | NVDA addon development, APIs, manifest, store submission | NVDA Addon Specialist | | Building a11y scanning tools, rule engines, parsers | Accessibility Tool Builder | | Desktop a11y APIs, UIA, MSAA, NSAccessibility | Desktop A11y Specialist | | Screen reader testing, keyboard testing, AT verification | Desktop A11y Testing Coach | | Web accessibility auditing | Web Accessibility Wizard | | Document accessibility auditing | Document Accessibility Wizard |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,820 | 13,357 | -10% | 1 | 1 | 0% | 2,644 | 2,602 | -2% | 0 | 0 | — |
case-02 | fail→pass | 8,599 | 7,415 | -14% | 1 | 1 | 0% | 1,509 | 1,790 | +19% | 0 | 0 | — |
case-03 | pass→pass | 13,379 | 11,953 | -11% | 1 | 1 | 0% | 2,555 | 2,483 | -3% | 0 | 0 | — |
case-04 | pass→pass | 13,327 | 11,098 | -17% | 1 | 1 | 0% | 2,450 | 2,463 | +1% | 0 | 0 | — |
case-05 | fail→pass | 12,028 | 4,403 | -63% | 1 | 1 | 0% | 1,981 | 1,141 | -42% | 0 | 0 | — |
case-06 | pass→pass | 10,373 | 5,959 | -43% | 1 | 1 | 0% | 1,571 | 1,449 | -8% | 0 | 0 | — |
case-07 | fail→fail | 8,886 | 5,871 | -34% | 1 | 1 | 0% | 1,484 | 1,437 | -3% | 0 | 0 | — |
case-08 | fail→pass | 15,230 | 5,664 | -63% | 1 | 1 | 0% | 2,272 | 1,301 | -43% | 0 | 0 | — |
case-09 | pass→pass | 9,673 | 4,066 | -58% | 1 | 1 | 0% | 1,449 | 1,126 | -22% | 0 | 0 | — |
case-10 | pass→pass | 9,910 | 6,097 | -38% | 1 | 1 | 0% | 1,524 | 1,478 | -3% | 0 | 0 | — |
case-11 | fail→pass | 9,377 | 4,241 | -55% | 1 | 1 | 0% | 1,491 | 1,054 | -29% | 0 | 0 | — |
case-12 | fail→pass | 10,082 | 3,400 | -66% | 1 | 1 | 0% | 1,596 | 1,012 | -37% | 0 | 0 | — |
case-13 | pass→pass | 11,624 | 5,168 | -56% | 1 | 1 | 0% | 1,859 | 1,308 | -30% | 0 | 0 | — |
case-14 | pass→pass | 8,134 | 5,878 | -28% | 1 | 1 | 0% | 1,403 | 1,409 | +0% | 0 | 0 | — |
case-15 | pass→pass | 2,547 | 1,781 | -30% | 1 | 1 | 0% | 423 | 745 | +76% | 0 | 0 | — |
case-16 | pass→fail | 7,890 | 4,835 | -39% | 1 | 1 | 0% | 1,298 | 1,260 | -3% | 0 | 0 | — |
case-17 | fail→pass | 12,505 | 8,738 | -30% | 1 | 1 | 0% | 2,318 | 2,079 | -10% | 0 | 0 | — |
case-18 | pass→pass | 9,954 | 3,447 | -65% | 1 | 1 | 0% | 1,663 | 1,009 | -39% | 0 | 0 | — |
case-19 | fail→pass | 24,843 | 11,888 | -52% | 1 | 1 | 0% | 2,638 | 2,665 | +1% | 0 | 0 | — |
case-20 | pass→fail | 11,188 | 7,229 | -35% | 1 | 1 | 0% | 2,106 | 1,824 | -13% | 0 | 0 | — |
case-21 | fail→pass | 6,015 | 5,874 | -2% | 1 | 1 | 0% | 968 | 1,412 | +46% | 0 | 0 | — |
case-22 | pass→pass | 11,462 | 4,974 | -57% | 1 | 1 | 0% | 1,625 | 1,253 | -23% | 0 | 0 | — |
case-23 | pass→pass | 14,475 | 12,746 | -12% | 1 | 1 | 0% | 2,147 | 2,385 | +11% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are 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.