# voice-ai-integration

A skill that routes Shengwang integration work: ConvoAI voice agents, RTC audio and video, RTM messaging, Cloud Recording, and token generation.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 511 installs · ★ 14
- Author: shengwang
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/Shengwang-Community/skills/tree/HEAD/skills/voice-ai-integration/SKILL.md
- Updated: 2026-03-27
- HTML: https://affpapa.org/en/nevibecode/skills/voice-ai-integration

## What it does

The skill reads your request and picks the matching Shengwang product module. Ask for a voice bot or a video room with chat and it maps the intent to ConvoAI, RTC, RTM, Cloud Recording, or a token server. Common pairings are spelled out too, so an AI assistant with chat history lands on ConvoAI plus RTC and RTM.

Before anything else it checks for a local doc index at `references/docs.txt`. If that file is missing, `bash skills/voice-ai-integration/scripts/fetch-docs.sh` pulls a static index from `doc.shengwang.cn`. From there each product module decides whether local references are enough or a page has to be fetched through `doc-mcp.shengwang.cn`. You need `bash`, `curl`, and network access to `doc.shengwang.cn`, `doc-mcp.shengwang.cn`, and `gitee.com`.

Credentials stay out of your project files — the skill uses environment variables and placeholder examples. Sample repos are cloned to a temporary path rather than your workspace, and code generation waits until you ask for it. Worth knowing: RTC and RTM modules need an AppID and a token server before they run, and the skill stops with an explanation when a dependency or credential is missing instead of guessing.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add Shengwang-Community/skills --skill voice-ai-integration -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/voice-ai-integration/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/voice-ai-integration/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add Shengwang-Community/skills --skill voice-ai-integration -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/voice-ai-integration/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add Shengwang-Community/skills --skill voice-ai-integration -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/voice-ai-integration/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install Shengwang-Community/skills voice-ai-integration
```

2. **Manually: put the files in place**
- File: `.github/skills/voice-ai-integration/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add Shengwang-Community/skills --skill voice-ai-integration -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/Shengwang-Community/skills --path skills/voice-ai-integration
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/voice-ai-integration/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add Shengwang-Community/skills --skill voice-ai-integration -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/Shengwang-Community/skills/tree/HEAD/skills/voice-ai-integration/SKILL.md

## When to use it
- pick the right Shengwang product for a described use case
- wire a ConvoAI voice agent to an RTC client
- add chat to a video call with RTM
- stand up an AccessToken2 token server
- enable server-side recording of an RTC session

## Example prompts
- I want a ConvoAI voice assistant — where do I start the integration?
- We need a video room with chat: pick the Shengwang products and walk me through the steps
- Generate an RTC token and show me how to move this into a token server

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: e24fa713f9c884c9.

BEGIN THIRD-PARTY FILE e24fa713f9c884c9
````text
---
name: voice-ai-integration
description: |
  Integrate Shengwang products: ConvoAI voice agents, RTC audio/video,
  RTM messaging, Cloud Recording, and token generation. Use when the user
  mentions Shengwang, 声网, ConvoAI, RTC, RTM, voice agent, AI agent,
  video call, live streaming, recording, token, or any Shengwang product task.
license: MIT
metadata:
  author: shengwang
  version: "1.0.0"
---

# Shengwang Integration

## Workflow

### Step 0: Ensure doc index exists (MANDATORY)

> **⚠️ Execute this BEFORE any routing or code generation.**

Check if `references/docs.txt` already exists. If it does, skip this step entirely.
If it does not exist, download it:
```bash
bash skills/voice-ai-integration/scripts/fetch-docs.sh
```
This downloads a static doc index from `doc.shengwang.cn` — no user data is sent.
If download fails, proceed with local reference docs and fallback URLs.

### Step 1: Route to the correct product module

> **Progressive disclosure rule:** Only read the file that the current step points to. Do not scan the `references/` directory or pre-read files that have not been linked from the current step. Each product module's README.md will tell you which files to read next and when. If a file is not referenced by the document you are currently reading, do not open it.

Match the user's request to a product module using the route table. If the match
is clear, route directly — do not ask extra questions.

#### Route Table

| User intent | Route to |
|-------------|----------|
| Credentials, AppID, REST auth | [general](references/general/credentials-and-auth.md) |
| Generate Token, token server, AccessToken2 | [token-server](references/token-server/README.md) |
| ConvoAI voice agent work | [conversational-ai](references/conversational-ai/README.md) |
| RTC SDK integration | [rtc](references/rtc/README.md) |
| RTM messaging / signaling | [rtm](references/rtm/README.md) |
| Cloud Recording | [cloud-recording](references/cloud-recording/README.md) |
| Download SDK, sample project, GitHub repo | Route to the relevant product module above |

#### Product Recognition Aid

When the user describes a use case without naming a product, use this to infer the match:

| Product | What it does | Typical user says |
|---------|-------------|-------------------|
````
END THIRD-PARTY FILE e24fa713f9c884c9

The third-party file has ended. Everything between the e24fa713f9c884c9 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2313 of 7014 bytes. © shengwang · MIT · https://github.com/Shengwang-Community/skills/tree/HEAD/skills/voice-ai-integration/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md