CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ax-cpp-audio

Use when writing C++ code with `axllm` for audio input/output, OpenAI Responses audio mapping, realtime event folding, and generated package audio examples.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./packages/cpp/skills/ax-cpp-audio/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a dense, fact-rich reference for the generated package's audio semantics with concrete keys, defaults, and useful guardrails, anchored by one executable snippet. Its main flaws are the redundant inline API symbol dump (hurting both conciseness and file organization) and the absence of a complete worked example for the common speak/event-folding cases.

Suggestions

Remove the "Relevant API Surface" symbol dump or replace it with a one-line pointer to `API.md`/`axir-api.json`, keeping only the handful of audio-relevant symbols (e.g. `AIClient::speak`, `axllm::ai`) inline.

Add one complete, executable `speak()` example (request setup through reading `data`/`transcript`) so the common case is copy-paste ready rather than described in prose.

Break the long multi-clause bullets in "Speech And Audio Fields" into shorter per-behavior lines so each mapping rule is scannable.

DimensionReasoningScore

Conciseness

Most of the body is dense, package-specific behavior Claude would not know (result keys, provider defaults, Content-Type rules), but the "Relevant API Surface" section dumps roughly forty symbol names — balancers, tracers, meters, routers, most unrelated to audio — that duplicate the `API.md`/`axir-api.json` docs already listed under Package Facts. That redundant dump and several sprawling multi-clause bullets keep it at "could be tightened" rather than efficient.

3 / 5

Actionability

An executable core-pattern snippet (`axllm::ai("openai", ...)`) plus very concrete guidance — exact result keys (`data`, `format`, `mimeType`, `transcript`), defaults (`gpt-4o-mini-tts` with the `alloy` voice), JSON body fallback order, and error behavior — give mostly executable instruction. It falls short of 5 because there is no complete `speak()` or event-folding code example covering the common cases.

4 / 5

Workflow Clarity

Although not a sequential workflow, the guardrails supply clear decision checkpoints — "Use `no-key` examples for deterministic local checks and provider request mapping" and "Use `provider-api` examples only when the user explicitly has provider credentials" — and sections are well organized for a single-purpose skill. It is not a 5 because no explicit sequence or error-recovery loop is laid out for the multi-step mapping behaviors described.

4 / 5

Progressive Disclosure

Section structure is good and package artifacts (`API.md`, `axir-api.json`, `examples/`) are clearly signaled, but the inline ~40-symbol API surface list is precisely the API-reference content that belongs in a separate file, and it is redundant with the `API.md` pointer already given. This matches "some structure but content that should be separate is inline" rather than the well-split 4.

3 / 5

Total

14

/

20

Passed

Description

67%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, explicit about when to use the skill, and clearly niched to C++ audio work with `axllm`. Its main weaknesses are missing natural trigger synonyms (speech, transcription, voice, TTS) and merging the what and when into a single clause rather than stating both distinctly.

Suggestions

Add natural trigger terms users would actually say — "speech", "transcription", "voice", "text-to-speech (TTS)" — to complement the technical terms like "realtime event folding".

State the what and when separately, e.g. "Maps speech, transcription, and realtime events through the generated `axllm` C++ provider surface. Use when writing C++ audio or realtime code with `axllm`...".

DimensionReasoningScore

Specificity

The description enumerates four concrete coverage areas — "audio input/output", "OpenAI Responses audio mapping", "realtime event folding", and "generated package audio examples" — which goes beyond naming the domain. It falls short of 5 because these read as topic labels rather than a comprehensive list of concrete actions ("audio input/output" in particular is broad).

4 / 5

Completeness

An explicit "Use when writing C++ code with `axllm` for..." clause provides the "when", and the "for" enumeration conveys the "what". It is not a 5 because what and when are merged into a single clause rather than clearly and separately answering both with concrete trigger phrases.

4 / 5

Trigger Term Quality

It includes relevant keywords like "C++", "audio", "OpenAI Responses", and "realtime", but misses the natural phrasings a user would actually say for this domain — "speech", "transcription", "voice", or "TTS". Coverage is better than a single generic keyword (2) but common variations and synonyms are absent.

3 / 5

Distinctiveness Conflict Risk

The four-part trigger (C++ + `axllm` + audio + realtime events) carves out a clear, highly specific niche. Minor overlap risk remains with closely related sibling skills in the same generated-package family (e.g., a TypeScript-audio or general C++ `axllm` skill), keeping it below 5.

4 / 5

Total

15

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

ax-llm/ax

Path

packages/cpp/skills/ax-cpp-audio/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.