Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when writing Go code with `github.com/ax-llm/ax/packages/go` for audio input/output, OpenAI Responses audio mapping, realtime event folding, and generated package audio examples.
.claude/skills/ax-llm-ax-go-audio/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 29 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -16% | 0% |
This skill helps an agent write Go code with the generated Ax package github.com/ax-llm/ax/packages/go. Use the generated package API, examples, and manifests; do not import TypeScript-only APIs unless you are editing the TypeScript package.
github.com/ax-llm/ax/packages/go.API.md and axir-api.json.axir-capabilities.json.examples/.javascript-goja.goimport ax "github.com/ax-llm/ax/packages/go" llm := ax.NewAI("openai", map[string]ax.Value{"apiKey": os.Getenv("OPENAI_API_KEY")})
axllm.NewAI, axllm.Typesafe, axllm.AxAITypesafeClient, context.Context, axllm.AxAIServiceAbortedError, axllm.GetSupportedAIModels, axllm.AxCredentialRequest, axllm.AxCredentialProvider, AxOwnedClientFactory.OwnedWorkerFactory, axllm.AxChatSession, axllm.AxChatStream, axllm.OpenAICompatibleClient, axllm.OpenAIResponsesClient, axllm.GoogleGeminiClient, axllm.AnthropicClient, axllm.AxUsageContext, axllm.AxUsageEvent, axllm.AxUsageObserver, axllm.SetUsageObserver, axllm.AxRuntimeHooks, axllm.AxRateLimitInfo, axllm.AxRateLimiter, axllm.AxTracer, axllm.AxMeter, axllm.AxGlobals, axllm.SetRateLimiter, axllm.SetTracer, axllm.SetMeter, axllm.AxBalancer, axllm.AxBalancerAdaptiveStrategy, axllm.AxBalancerStatsStore, axllm.AxInMemoryBalancerStatsStore, axllm.CreateBalancerRouteStats, axllm.UpdateBalancerRouteStats, axllm.SampleBalancerRouteHealth, axllm.MultiServiceRouter, axllm.ProviderRouterprovider-api examples only when the user explicitly has provider credentials available.no-key examples for deterministic local checks and provider request mapping.tools/*/skills/ into user packages.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,499 | 14,668 | +17% | 1 | 1 | 0% | 2,297 | 2,157 | -6% | 0 | 0 | — |
case-02 | fail→pass | 21,427 | 17,882 | -17% | 1 | 1 | 0% | 3,478 | 4,151 | +19% | 0 | 0 | — |
case-03 | fail→pass | 19,926 | 11,480 | -42% | 1 | 1 | 0% | 3,594 | 2,984 | -17% | 0 | 0 | — |
case-04 | fail→fail | 9,057 | 2,836 | -69% | 1 | 1 | 0% | 1,607 | 1,130 | -30% | 0 | 0 | — |
case-05 | fail→pass | 14,003 | 4,736 | -66% | 1 | 1 | 0% | 2,336 | 1,352 | -42% | 0 | 0 | — |
case-06 | fail→pass | 19,389 | 3,987 | -79% | 1 | 1 | 0% | 1,565 | 1,307 | -16% | 0 | 0 | — |
case-07 | pass→pass | 12,962 | 4,986 | -62% | 1 | 1 | 0% | 1,763 | 1,234 | -30% | 0 | 0 | — |
case-08 | pass→pass | 24,244 | 3,902 | -84% | 1 | 1 | 0% | 2,312 | 1,257 | -46% | 0 | 0 | — |
case-09 | pass→pass | 8,128 | 5,599 | -31% | 1 | 1 | 0% | 1,123 | 1,578 | +41% | 0 | 0 | — |
case-10 | pass→pass | 14,195 | 6,458 | -55% | 1 | 1 | 0% | 1,792 | 1,578 | -12% | 0 | 0 | — |
case-11 | pass→pass | 14,613 | 13,188 | -10% | 1 | 1 | 0% | 2,514 | 2,777 | +10% | 0 | 0 | — |
case-12 | pass→pass | 25,010 | 10,774 | -57% | 1 | 1 | 0% | 2,237 | 2,452 | +10% | 0 | 0 | — |
case-13 | pass→fail | 24,309 | 7,157 | -71% | 1 | 1 | 0% | 2,647 | 1,551 | -41% | 0 | 0 | — |
case-14 | fail→pass | 14,535 | 4,777 | -67% | 1 | 1 | 0% | 2,482 | 1,406 | -43% | 0 | 0 | — |
case-15 | fail→pass | 7,559 | 3,474 | -54% | 1 | 1 | 0% | 1,060 | 1,199 | +13% | 0 | 0 | — |
case-16 | fail→pass | 20,339 | 3,464 | -83% | 1 | 1 | 0% | 2,779 | 1,214 | -56% | 0 | 0 | — |
case-17 | fail→pass | 17,809 | 3,675 | -79% | 1 | 1 | 0% | 2,832 | 1,115 | -61% | 0 | 0 | — |
case-18 | pass→pass | 11,757 | 4,141 | -65% | 1 | 1 | 0% | 1,972 | 1,212 | -39% | 0 | 0 | — |
case-19 | fail→pass | 28,079 | 2,373 | -92% | 1 | 1 | 0% | 4,155 | 919 | -78% | 0 | 0 | — |
case-20 | fail→pass | 24,646 | 2,384 | -90% | 1 | 1 | 0% | 3,589 | 920 | -74% | 0 | 0 | — |
case-21 | pass→pass | 12,576 | 5,335 | -58% | 1 | 1 | 0% | 1,822 | 1,308 | -28% | 0 | 0 | — |
case-22 | fail→pass | 15,041 | 4,621 | -69% | 1 | 1 | 0% | 2,031 | 1,299 | -36% | 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 +50 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 9/2/2026 | +82% |
| gemini-3.6-flash | verified | 8/27/2026 | +57% |
| gemini-3.6-flash | verified | 8/18/2026 | +64% |
| gemini-3.6-flash | verified | 8/10/2026 | +55% |
Other measured skills in the registry, with their headline benchmark lift.