Getting Started

AskQuorum AI is a multi-model AI platform with a personal AI assistant (Kavya), a CLI for terminal-based workflows, and multi-channel support across WhatsApp, Telegram, and Discord.

Quick start: Install the CLI, configure your channels, and you're live in under 5 minutes.

Installation

Prerequisites

Node.js 18+ and npm
A supported OS (macOS, Linux, Windows via WSL)

Install via npm

npm install -g @askquorum-ai/cli

Verify installation

askquorum --version
askquorum help

Configuration

AskQuorum uses a YAML configuration file. On first run, a default config is created at ~/.askquorumai/config.yaml.

Core settings

# ~/.askquorumai/config.yaml
gateway:
  port: 8080
  host: localhost

ai:
  default_model: anthropic/claude-sonnet-4-20250514
  thinking: auto

channels:
  whatsapp:
    enabled: true
  telegram:
    enabled: false
  discord:
    enabled: false

Environment variables

API keys and secrets can be set via environment variables or stored in the encrypted credential vault:

export ANTHROPIC_API_KEY=sk-ant-...
export OPENAI_API_KEY=sk-...

Security: Never commit API keys to version control. Use the built-in vault (askquorum secrets set) for production.

WhatsApp Channel

Connect your AI assistant to WhatsApp using the Baileys library. The CLI manages session state and QR code authentication automatically.

Setup

askquorum channel add whatsapp
askquorum gateway start

Scan the QR code with your WhatsApp app to authenticate. The session persists across restarts.

Features

Text, image, and voice message support
Inline reply context
Group chat participation
OTP-based user verification

Telegram Channel

Connect via Telegram Bot API with a bot token from @BotFather.

Setup

askquorum channel add telegram --token YOUR_BOT_TOKEN

Features

Text, photo, voice, and document support
Inline keyboards and commands
Group and private chat modes

Discord Channel

Deploy as a Discord bot using a bot token from the Discord Developer Portal.

Setup

askquorum channel add discord --token YOUR_BOT_TOKEN

Features

Multi-server support
Slash commands and message reactions
Thread and DM support
Role-based access control

Plugins

Extend AskQuorum with plugins for additional capabilities. Plugins are Node.js modules that hook into the gateway lifecycle.

Available plugins

calendar — Google Calendar integration
email — Gmail reading and sending
web-search — Brave Search API integration
tts — Text-to-speech via ElevenLabs
browser — Headless browser automation

Enable a plugin

# In config.yaml
plugins:
  calendar:
    enabled: true
  email:
    enabled: true
    provider: gmail

Creating custom plugins

Plugins export an init() function that receives the gateway context:

// my-plugin/index.js
module.exports = {
  name: 'my-plugin',
  init(ctx) {
    ctx.on('message', async (msg) => {
      // Handle messages
    });
  }
};

CLI Reference

Gateway commands

askquorum gateway start     # Start the gateway daemon
askquorum gateway stop      # Stop the gateway
askquorum gateway restart   # Restart the gateway
askquorum gateway status    # Check gateway status

Channel commands

askquorum channel list              # List configured channels
askquorum channel add <type>        # Add a channel (whatsapp|telegram|discord)
askquorum channel remove <type>     # Remove a channel

Secret management

askquorum secrets list              # List stored secrets
askquorum secrets set <key> <val>   # Store an encrypted secret
askquorum secrets delete <key>      # Delete a secret

Utility commands

askquorum --version     # Print version
askquorum help          # Show help
askquorum doctor        # Diagnose configuration issues

API Reference

The REST API reference is coming soon. The gateway exposes endpoints for message handling, user management, and plugin orchestration on localhost:8080.

Base URL

http://localhost:8080/api/

Endpoints (preview)

POST /api/verify/lookup — Look up a user by phone number
POST /api/verify/send-otp — Send OTP for verification
POST /api/verify/confirm — Confirm OTP
GET /api/health — Health check

Full API documentation with request/response schemas will be published here.