voicelive-realtime

voicelive-realtime is a skill for Cursor from JinLee794/agent-framework-skills. It costs 101 tokens per session (4,345 once invoked), scanned A, original, MIT.

A guide for building real-time voice applications with the Azure AI VoiceLive software kit. It covers setup choices, microphone and speaker prototypes, live conversations, audio formats, voices, and tool use.

In plain words
What is it for?
Use it when connecting an application to VoiceLive, choosing between an Azure-managed model and your own hosted model, or building a local two-way voice prototype.
Why use it?
It helps avoid configuration mistakes that can make a connection fail for a reason that is hard to identify. It also provides a checked path for testing audio input, output, interruptions, and shutdown.

Skill for Cursor

Written for Cursor: installed under .cursor/.

Good fit Use it when connecting an application to VoiceLive, choosing between an Azure-managed model and your own hosted model, or building a local two-way voice prototype.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jinlee794/agent-framework-skills/voicelive-realtime
Install

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.

Any agent
npx skills add JinLee794/agent-framework-skills --skill voicelive-realtime
Clone the repo
git clone --depth 1 https://github.com/JinLee794/agent-framework-skills

Made for: Cursor.

Wrote 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.

agentmods badge for voicelive-realtime

README.md
[![agentmods](https://agentmods.dev/badge/skills/jinlee794/agent-framework-skills/voicelive-realtime/github.svg)](https://agentmods.dev/skills/jinlee794/agent-framework-skills/voicelive-realtime)
Your own site
<a href="https://agentmods.dev/skills/jinlee794/agent-framework-skills/voicelive-realtime"><img src="https://agentmods.dev/badge/skills/jinlee794/agent-framework-skills/voicelive-realtime/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.

agentmods 80×15 button for voicelive-realtime

Your own site · 80×15
<a href="https://agentmods.dev/skills/jinlee794/agent-framework-skills/voicelive-realtime"><img src="https://agentmods.dev/badge/skills/jinlee794/agent-framework-skills/voicelive-realtime.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 101 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,345 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00101 $0.04345
Opus 5 $0.00051 $0.02173
Sonnet 5 $0.00020 $0.00869
Haiku 4.5 $0.00010 $0.00434

Measured 10d ago against content hash ce09510ccab8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

voicelive-realtime 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 10d 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.

.cursor/skills/voicelive-realtime/SKILL.md · 417 lines

How it starts

The opening of the file, as written. The whole thing — 417 lines — stays where its author put it; the contents beside it link to each section on GitHub.

VoiceLive Realtime — GA Patterns

Start here

Any task that touches the VoiceLive connection or session begins with the configuration interview below. Do not edit .env, the voice section of config/voice/<name>.yaml, or src/<package>/voice/ until the user has confirmed the summary — a wrong model source fails at connect() with an error that looks nothing like its cause.

For a runnable local full-duplex prototype using the device microphone and speakers, follow references/local-audio-prototype.md. It isolates the audio transport before adding the MAF brain and includes device preflight, threading, barge-in, shutdown, and acceptance checks.

Configuration interview

Ask these in order, one question per turn, using the editor's question tool so the user can pick from options. Carry forward answers already stated in the conversation or already present in .env — state what you found and ask for confirmation instead of re-asking blind. Stop and ask whenever an answer would otherwise be inferred.

Q1 — Model source

VoiceLive can run a model it hosts for you, or a model deployment you own. Which one?

  • Voice Live-managed — pick from a pre-deployed catalogue. No deployment, no capacity planning, no quota to arrange. Fastest way to a working call.
  • Foundry-hosted (BYOM) — VoiceLive calls a deployment in your Foundry resource. Required for fine-tuned or provisioned-throughput deployments, for any model outside the managed catalogue, and to reuse the exact deployment the local MAF brain already calls.

This repo's .env.example ships BYOM because its FOUNDRY_MODEL is not in the managed catalogue. That is a default, not a constraint — say so rather than treating BYOM as settled.

Q2 — Model identifier

Load references/model-selection.md now and present the branch matching Q1. Do not recite the whole catalogue; narrow it first.

Read the full file on GitHub · 417 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 10d ago First seen · 417 lines · 101 tokens per session scan A ce09510ccab8

Subscribe to this mod's changes

voicelive-realtime is a skill published in the GitHub repository JinLee794/agent-framework-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 101 tokens to every session and 4,345 once invoked, about $0.0005 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories