Quick Voice

Spin up an instant browser voice session (OpenAI Realtime gpt-realtime-2) to close a topic in a short conversation instead of working through documents. Generic & white-label - works for any process. Supports live data work (read/update files, JSON, run commands), and distill mode (no tools, ends with a structured deliverable). Has a generic canvas that can display images, markdown, code, html, json, video, audio - perfect for "let's go over X" flows where the agent shows you items one by one and you react in real time. Use when user says "let's close this in a voice call", "run a quick voice

51 stars 14 forksImages, Video & Audio Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add aviz85/claude-skills-library

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add aviz85/claude-skills-library --skill quick-voice
Manual install
git clone --depth 1 https://github.com/aviz85/claude-skills-library.git /tmp/claude-skills-library
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills-library/plugins/quick-voice/skills/quick-voice ~/.claude/skills/quick-voice

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# quick-voice

White-label voice session generator. Each invocation produces a per-session web app that opens a WebRTC voice channel to OpenAI Realtime (`gpt-realtime-2`) with **context-specific instructions, tools, and canvas behavior**.

## When to use

- "תפעיל איתי שיחה קולית על X" / "let's have a quick voice call about X"
- Going through a list of items where speaking is faster than reading + clicking
- Closing a topic that needs a few decisions + updates (not a long-form plan)
- Producing a structured output (decisions / action items / notes) from a free-form discussion

## Two modes

| Mode | Tools | Output |
|---|---|---|
| **live** | File / JSON / bash tools available — agent updates real data during the call | A summary of changes made (in `output.md`) |
| **distill** | No data-mutation tools; canvas + `save_note` + `end_session` only | A long structured deliverable (decisions, notes, action items) in `output.md` |

The canvas is available in **both** modes.

## How to run a session

### Step 1 — figure out context

Look at the conversation. The user said one of:
- **Explicit:** "let's close the freelancer reviews" → topic = freelancer reviews
- **Implicit:** earlier in the conversation we generated 10 images → topic = "go over generated images"
- **Vague:** "תפעיל שיחה קולית" → call AskUserQuestion (single Q) to clarify topic + mode

### Step 2 — pick a runtime directory + generate the session config
Read the full skill on GitHub

Details

Skill name
quick-voice
Path
plugins/quick-voice/skills/quick-voice/SKILL.md
Repo updated
8 July 2026

FAQ

What does the Quick Voice skill do?

Spin up an instant browser voice session (OpenAI Realtime gpt-realtime-2) to close a topic in a short conversation instead of working through documents. Generic & white-label - works for any process. Supports live data work (read/update files, JSON, run commands), and distill mode (no tools, ends with a structured deliverable). Has a generic canvas that can display images, markdown, code, html, json, video, audio - perfect for "let's go over X" flows where the agent shows you items one by one and you react in real time. Use when user says "let's close this in a voice call", "run a quick voice

How do I install the Quick Voice skill?

Claude Code plugin: /plugin marketplace add aviz85/claude-skills-library. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Quick Voice popular?

It lives in aviz85/claude-skills-library, which has 51 GitHub stars and 14 forks.

Related skills

Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.