jarvis-mcp - GitLite

KO | EN

gitlite — search

#hacktoberfest #ai #golang #go #machine-learning #deep-learning #macos #nodejs #java #windows #cli #kubernetes #mcp #open-source

jarvis-mcp

(★ 94)

Bring your AI to life—talk to assistants instantly in your browser. Zero hassle, No API keys, No Whisper

File Explorer

Download Latest Version (.zip)

- - voice-interface.js
- - index.ts
  - types.ts
  - voice-logger.ts
  - voice-service.ts
- build.sh
- example.env
- package-lock.json
- package.json
- README.md
- tsconfig.json
- - marked.umd.js
  - purify.min.js
- app.js
- audio-buffer-to-wav.js
- index.html
- qr.js
- browser-interface.ts
- certificate-manager.ts
- cli.ts
- index.ts
- voice-queue.ts
.gitignore
config.md
create-bundle.cjs
package-lock.json
package.json
README.md
test-goodbye.cjs
test-https.js
test-pause-feature.html
test-pause-logic.js
test-plugin-export.cjs
test-plugin-https.cjs
test-plugin.cjs
test-speech-interruption.sh
test-user-interruption.sh
tsconfig.json

# Installation Guide

1. Get the code

git clone https://github.com/shantur/jarvis-mcp

Downloads the entire project code from GitHub to your computer.

cd jarvis-mcp

Moves into the project folder you just downloaded.

2. Official Install Script

Easy Recommended

Prerequisites

Node.js Node.js must be installed to use npm.

npm install -g @shantur/jarvis-mcp

Installs the package published on the npm registry globally — no need to clone the source.

✅ After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.

Pulled directly from this repo's README.

3. Node.js

Easy

Prerequisites

Git Needed to download the project code from GitHub.
Node.js Node.js must be installed to use npm. The LTS version is recommended.

npm install -g @shantur/jarvis-mcp

Installs the package published on the npm registry globally — no need to clone the source.

npm install && npm run build && npm start

Downloads and installs the libraries listed in package.json.

npm install

Downloads and installs the libraries listed in package.json.

npm run build

Builds optimized production files.

npm run dev # Watch mode

Starts the development/run server.

✅ After running the command, open the address shown in the terminal (usually something like http://localhost:3000) in your browser.

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Jarvis MCP

Bring your AI to life—talk to assistants instantly in your browser. Compatible with Claude Desktop, OpenCode, and other MCP-enabled AI tools.

> ✅ No extra software, services, or API keys required—just open the web app in your browser and grant microphone access.

## Features

🎙️ **Voice Conversations** - Speak naturally with AI assistants  
🌍 **30+ Languages** - Speech recognition in multiple languages  
📱 **Remote Access** - Use from phone/tablet while AI runs on computer  
⚙️ **Smart Controls** - Collapsible settings, always-on mode, custom voices  
⏱️ **Dynamic Timeouts** - Intelligent wait times based on response length  
🧰 **Zero Extra Software** - Runs entirely in your browser—no extra installs or API keys  
🔌 **Optional Whisper Streaming** - Plug into a local Whisper server for low-latency transcripts

## Easy Installation

### 🚀 One-Command Setup

**Claude Desktop:**
```bash
npx @shantur/jarvis-mcp --install-claude-config
# Restart Claude Desktop and you're ready!
```

**OpenCode (in current project):**
```bash
npx @shantur/jarvis-mcp --install-opencode-config --local
npx @shantur/jarvis-mcp --install-opencode-plugin --local
# Start OpenCode and use the converse tool
```

**Claude Code CLI:**
```bash
npx @shantur/jarvis-mcp --install-claude-code-config --local
# Start Claude Code CLI and use voice tools
```

### 🤖 Why Install the OpenCode Plugin?

- Stream voice messages into OpenCode even while tools are running or tasks are in progress.
- Auto-forward pending Jarvis MCP conversations so you never miss a user request.
- Works entirely locally—no external services required, just your OpenCode project and browser.
- Installs with one command and stays in sync with the latest Jarvis MCP features.

### 📦 Manual Installation

**From NPM:**
```bash
npm install -g @shantur/jarvis-mcp
jarvis-mcp
```

**From Source:**
```bash
git clone <repository-url>
cd jarvis-mcp
npm install && npm run build && npm start
```

## How to Use

1. **Hook it into your AI tool** – Use the install command above for Claude Desktop, OpenCode, or Claude Code so the MCP server is registered.
2. **Kick off a voice turn** – Call the `converse` tool from your assistant; Jarvis MCP auto-starts in the background and pops open `https://localhost:5114` if needed.
3. **Allow microphone access** – Approve the browser prompt the first time it appears.
4. **Talk naturally** – Continue using `converse` for every reply; Jarvis MCP handles the rest.

### Voice Commands in AI Chat

```
Use the converse tool to start talking:
- converse("Hello! How can I help you today?", timeout: 35)
```

## Browser Interface

The web interface provides:

- **Voice Settings** (click ⚙️ to expand)
  - Language selection (30+ options)
  - Voice selection
  - Speech speed control
  - Always-on microphone mode
  - Silence detection sensitivity & timeout (for Whisper streaming)
- **Smart Controls**
  - Pause during AI speech (prevents echo)
  - Stop AI when user speaks (natural conversation)
- **Mobile Friendly** - Works on phones and tablets

## Remote Access

Access from any device on your network:

1. Find your computer's IP: `ifconfig | grep inet` (Mac/Linux) or `ipconfig` (Windows)
2. Visit `https://YOUR_IP:5114` on your phone/browser
3. Accept the security warning (self-signed certificate)
4. Grant microphone permissions

Perfect for continuing conversations away from your desk!

## Configuration

### Environment Variables

```bash
export MCP_VOICE_AUTO_OPEN=false  # Disable auto-opening browser
export MCP_VOICE_HTTPS_PORT=5114  # Change HTTPS port
export MCP_VOICE_STT_MODE=whisper  # Switch the web app to Whisper streaming
export MCP_VOICE_WHISPER_URL=http://localhost:12017/v1/audio/transcriptions  # Whisper endpoint (full path)
export MCP_VOICE_WHISPER_TOKEN=your_token  # Optional Bearer auth for Whisper server
```

### Whisper Streaming Mode

- Whisper mode records raw PCM in the browser, converts it to 16 kHz mono WAV, and streams it through the built-in HTTPS proxy, so the local `whisper-server` sees OpenAI-compatible requests.
- By default we proxy to the standard `whisper-server` endpoint at `http://localhost:12017/v1/audio/transcriptions`; point `MCP_VOICE_WHISPER_URL` at your own host/port if you run it elsewhere.
- The UI keeps recording while transcripts are in flight and ignores Whisper’s non-verbal tags (e.g. `[BLANK_AUDIO]`, `(typing)`), so only real speech is queued.
- To enable it:
  1. Run your Whisper server locally (e.g. `whisper-server` from `pfrankov/whisper-server`).
  2. Set the environment variables above (`MCP_VOICE_STT_MODE=whisper` and the full `MCP_VOICE_WHISPER_URL`).
  3. Restart `jarvis-mcp` and hard-refresh the browser (empty-cache reload) to load the streaming bundle.
  4. Voice status (`voice_status()` tool) now reports whether Whisper or browser STT is active.

### Ports

- **HTTPS**: 5114 (required for microphone access)
- **HTTP**: 5113 (local access only)

## Requirements

- Node.js 18+
- Google Chrome (only browser tested so far)
- Microphone access
- Optional: Local Whisper server (like `pfrankov/whisper-server`) if you want streaming STT via `MCP_VOICE_STT_MODE=whisper`

## Troubleshooting

**Certificate warnings on mobile?**
- Tap "Advanced" → "Proceed to site" to accept self-signed certificate

**Microphone not working?**
- Ensure you're using HTTPS (not HTTP)
- Check browser permissions
- Try refreshing the page

**AI not responding to voice?**
- Make sure the `converse` tool is being used (not just `speak`)
- Check that timeouts are properly calculated

## Development

```bash
npm install
npm run build
npm run dev     # Watch mode
npm run start   # Run server
```

## License

MIT