Name: voice-ai-development
Rating: 46.400000000000006 (1 reviews)
Author: administrakt0r

voice-ai-development

You are an expert in building real-time voice applications. You think in terms of latency budgets, audio quality, and user experience. You know that voice apps feel magical when fast and broken when slow.

Quality

49%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/voice-ai-development/SKILL.md

The canonical home for this skill is voice-ai-development in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

50%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 content is a well-sectioned catalog of concrete voice-AI patterns with useful anti-patterns, but the flagship OpenAI code example is truncated, there is no sequenced workflow with validation, and everything is inlined in one file with no progressive disclosure. It is actionable for two of three providers but not fully copy-paste ready.

Suggestions

Complete the truncated OpenAI Realtime example (the event-handling loop ends mid-token at 'resp') so all code blocks are executable.

Add a short end-to-end streaming pipeline workflow with a validation/feedback checkpoint (e.g. verify barge-in and latency budget) to raise workflow clarity.

Move per-provider API reference detail into reference files (e.g. references/openai-realtime.md) and keep SKILL.md as an overview with one-level-deep links.

Dimension	Reasoning	Score
Conciseness	The body is mostly code and 'When to use' notes without padding Claude's prior knowledge, but repeats the persona prose from the description and carries inline explanatory comments plus capability/requirement lists that could be tightened, matching the mostly-efficient anchor.	3 / 5
Actionability	The Vapi and Deepgram+ElevenLabs examples are concrete and near-executable, but the lead OpenAI Realtime example is truncated mid-statement ('if event["type"] == "resp'), leaving a key pattern incomplete and pushing it to the incomplete-guidance anchor.	3 / 5
Workflow Clarity	Patterns are organized with 'When to use' decision guidance and anti-pattern 'Instead' remediation, but there is no sequenced multi-step workflow or validation checkpoint; this matches the anchor of structure present but checkpoints missing or implicit.	3 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ absent) and all provider detail is inlined in a single ~300-line file; sections are well-labeled, but content that could be split (per-provider reference) is inline with no one-level-deep references, matching the some-structure anchor.	3 / 5
	Total	12 / 20 Passed

Description

48%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 establishes a clear voice-AI niche but reads as a persona statement rather than a capability-and-trigger description: it is second person, lacks a 'Use when...' clause, and lists no concrete actions. It is distinguishable but not actionable as a trigger.

Suggestions

Rewrite in third person with concrete actions, e.g. 'Builds and optimizes real-time voice applications using OpenAI Realtime, Vapi, Deepgram, and ElevenLabs.'

Add an explicit trigger clause: 'Use when the user wants to build voice agents, speech-to-text/text-to-speech pipelines, or low-latency voice apps.'

Include natural synonyms and file/SDK terms users say (voice assistant, TTS, STT, speech, WebRTC) to improve trigger term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ('building real-time voice applications') and one framing action, but offers no concrete list of actions; the heavy second-person voice ('You are an expert', 'You think') triggers the rubric's -1 specificity penalty, dropping the base of 3 to 2.	2 / 5
Completeness	The 'what' is present but vague (building real-time voice apps framed as a mindset), and there is no 'Use when...' or equivalent trigger clause; the rubric caps completeness at 3 for a missing explicit when-guidance.	3 / 5
Trigger Term Quality	Phrases like 'real-time voice applications', 'voice apps', 'latency budgets', and 'audio quality' are relevant natural terms, but common synonyms a user might say (speech, TTS, STT, voice assistant, voice agent) are missing, matching the anchor for some relevant keywords with missing variations.	3 / 5
Distinctiveness Conflict Risk	'Real-time voice applications' is a clear niche distinct from most skills with minimal overlap risk, though the triggers are not explicit enough to reach the 5 anchor.	4 / 5
	Total	12 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/voice-ai-development/SKILL.md
Commit: 900c504

Reviewed: 5 hours ago

Table of Contents

Discovery Implementation Validation

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.