Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jnPiyush/AgentX --skill voice-agentsgit clone --depth 1 https://github.com/jnPiyush/AgentXWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jnpiyush/agentx/voice-agents)<a href="https://agentmods.dev/skills/jnpiyush/agentx/voice-agents"><img src="https://agentmods.dev/badge/skills/jnpiyush/agentx/voice-agents/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jnpiyush/agentx/voice-agents"><img src="https://agentmods.dev/badge/skills/jnpiyush/agentx/voice-agents.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Memory Poisoning · line 99 Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.Fix: Implement context-window management that detects and rejects padding or stuffing attempts. Prioritize system instructions over user-injected content.
- medium Excessive Agency · line 110 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00093 | $0.01323 |
| Opus 5.5 | $0.00037 | $0.00529 |
| Sonnet 5.5 | $0.00019 | $0.00265 |
| Haiku 4.5 | $0.00009 | $0.00132 |
Grade A, and why
voice-agents scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Voice Agents
Purpose: Build voice experiences that feel like a phone call, not a chatbot reading aloud.
When to Use This Skill
- Real-time voice assistants (web, mobile, kiosk)
- Telephony agents (inbound / outbound calls, IVR replacement)
- Multimodal copilots that speak and listen
- Accessibility features that require interruption / barge-in
Architecture Choice
Speech-to-Speech (S2S, single model)
audio in -> [Realtime model] -> audio out
Pros: <500ms latency, natural prosody, can hear tone
Cons: fewer model choices, higher cost, fewer guardrail hooks
Cascaded
audio in -> [STT] -> [LLM] -> [TTS] -> audio out
Pros: any LLM, full control, cheaper, easier guardrails
Cons: 800-1500ms latency, lost prosody, brittle endpointing
Pick S2S when latency and natural turn-taking dominate. Pick cascaded when you need a specific LLM, tighter cost control, or richer logging.
Provider Snapshot (April 2026)
| Provider | Mode | Notes |
|---|---|---|
| OpenAI Realtime / GPT Realtime | S2S | WebRTC, voice activity detection, function calling |
| Azure Voice Live | S2S | Foundry-integrated, Azure RAI controls |
| Gemini Live | S2S | Multimodal (audio + video), tool use |
| Deepgram Voice Agent | S2S | Telephony focus, low latency |
| ElevenLabs Conversational AI | S2S | Best TTS quality, broad voice catalog |
| Cartesia, Rime, Deepgram Aura | TTS | For cascaded pipelines |
| AssemblyAI, Deepgram Nova | STT | For cascaded pipelines |
| LiveKit Agents, Pipecat, Vocode | Orchestration | Wire the pipeline together |
Latency Budget
End-to-end target for natural conversation: <800ms mouth-close to mouth-open.
| Stage | Cascaded budget | S2S budget |
|---|---|---|
| Endpointing (silence detection) | 200-300ms | 100-200ms |
| STT first-token | 150-300ms | n/a |
| LLM first-token | 200-500ms | 200-500ms |
| TTS first-audio | 150-300ms | bundled |
| Network jitter | 50-150ms | 50-150ms |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed 4fbac3be3eee
- 1mo ago First seen · 140 lines · 93 tokens per session scan A 7f49271ff359
voice-agents is a skill published in the GitHub repository jnPiyush/AgentX (16 stars, last pushed 3d ago), licensed Apache-2.0. It adds 93 tokens to every session and 1,323 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
ml-kit-genai-prompt-api
Analyzes Android codebases to implement ML Kit GenAI Prompt API. Use this skill to send natural language requests on-device to Gemini Nano, use structured output with Prompt API, implement prefix caching, optimize the current prompt, or apply best practices.".
model-authoring
Empirical rules for authoring PyTorch models for on-device execution on Apple platforms, covering energy-efficient inference, scalable compute, and correctness testing. Use this skill whenever the user is writing, debugging, or reviewing PyTorch model code intended for on-device execution — even if they don't…
working-with-coreai
Use this skill whenever the user mentions coreai-torch, TorchConverter, coreai-build, AIModel, AIProgram, .aimodel, or wants to export/compile/run a PyTorch model on Apple silicon (iPhone, iPad, Mac). Also triggers for "deploy on device", "optimize for on-device performance", onboarding new models to Core AI, or…
litert-runtime
Configures and implements LiteRT (.tflite) on-device inference on Android in Kotlin using the CompiledModel API on CPU and GPU. Use this skill for any Android LiteRT or .tflite task, including Gradle setup, image classification or vision/audio/embedding helpers, TensorBuffer reuse, Bitmap preprocessing…
foundation-models-on-device
Apple FoundationModels framework for on-device LLM — text generation, guided generation with @Generable, tool calling, and snapshot streaming in iOS 26+.
axiom-ai
Use when implementing, testing, or evaluating ANY Apple Intelligence, on-device AI, or speech-to-text feature. Covers Foundation Models, @Generable, LanguageModelSession, Tool protocol, eval suites, model-as-judge scoring, SpeechTranscriber, CoreML.