Quick Start

Get your AI agent into a voice meeting in minutes. Pick the integration that fits your setup.

1. Get an API Key

Sign up at chamade.io/dashboard, then create an API key. It starts with chmd_ and is shown only once — store it securely.

2. Choose Your Integration

MCP Server (HTTP — recommended)

For Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP client that supports the Streamable HTTP transport. Drop the snippet below in your MCP config file (.mcp.json, claude_desktop_config.json, .cursor/mcp.json…) and restart the client.

            json
            
        
{
  "mcpServers": {
    "chamade": {
      "type": "http",
      "url": "https://mcp.chamade.io/mcp/",
      "headers": {
        "Authorization": "Bearer chmd_..."
      }
    }
  }
}

Claude Code only: also launch with claude --dangerously-load-development-channels server:chamade --continue on every session to receive push events (incoming calls, DMs) in real time. No flag needed for the tools themselves — they work immediately.

Legacy stdio-only clients (older MCP clients without Streamable HTTP): use the @chamade/mcp-server@3 stdio shim — it's a thin wrapper around mcp-remote that bridges stdio to the same hosted HTTP endpoint. See the full MCP setup guide for the shim config.

Full MCP setup guide →

REST API

For any language or framework. Direct HTTP calls — no MCP client needed. Works great for pure-backend agents, non-LLM orchestration, webhooks, and anything that speaks HTTP.

            bash
            
        
curl -X POST https://chamade.io/api/call \
  -H "X-API-Key: chmd_..." \
  -H "Content-Type: application/json" \
  -d '{"platform":"discord","meeting_url":"https://discord.com/channels/..."}'

Full API reference →

Voice calls — BYO audio WebSocket

For voice calls, MCP and REST give you control (join, leave, chat, state), but the raw audio flows on a separate WebSocket so your agent can plug its own STT/TTS stack (OpenAI Realtime, LiveKit Agents, Pipecat, Deepgram Voice Agent, ElevenLabs, Cartesia, Whisper cascade, etc.). POST /api/call returns an audio block that tells your host code where to connect and what PCM format to stream.

            json
            
        
{
  "call_id": "abc-123",
  "capabilities": ["audio_in", "audio_out", "read", "write"],
  "audio": {
    "stream_url": "wss://chamade.io/api/call/abc-123/stream",
    "format": "pcm_s16le",
    "sample_rate": 48000,
    "frame_duration_ms": 20,
    "direction": "bidirectional",
    "docs": "https://chamade.io/docs/api#websocket"
  }
}

Your agent's host code (not the LLM itself) opens the stream_url in parallel with MCP/REST, pipes binary PCM frames in both directions between the call and your chosen STT/TTS, and lets the LLM drive the conversation via text. Hosted STT/TTS (Chamade runs the speech layer for you) is currently beta-gated — contact [email protected] for supervised access.

Audio WebSocket protocol →

3. Connect Your Platforms

Some platforms require setup in the Dashboard before use:

Platform	Setup needed
Discord	Works out of the box (shared bot) or add your own bot
Microsoft Teams	Connect Microsoft account (OAuth)
Google Meet	Connect Google account (OAuth)
Zoom	No setup — just pass a meeting URL
Telegram	Works out of the box or add your own bot
WhatsApp	No setup — invite the bot
Slack	Install the Slack app
Nextcloud Talk	Install addon + connect
SIP / Phone	Activate a phone number or bring your own trunk