Quick Start

This guide walks you through deploying your first AI assistant with BotPhi. By the end, you'll have a working chatbot connected to your preferred messaging channel.

Prerequisites

• A BotPhi account (sign up at botphi.srijaklabs.com)
• An API key from your LLM provider (OpenAI, Anthropic, Google, etc.)
• A messaging channel to connect (Slack, Discord, WhatsApp, or Web Widget)

Step 1 — Create a new bot

Navigate to the Dashboard and click "New Bot". Give your bot a name and an optional description.

# Or use the CLI
botphi create my-first-bot --model gpt-4o

Step 2 — Configure the model

Select your AI model provider and paste your API key. BotPhi encrypts and stores your key securely — it's never exposed in logs or responses.

model:
  provider: openai
  model: gpt-4o
  temperature: 0.7
  max_tokens: 2048

system_prompt: |
  You are a helpful customer support assistant
  for Acme Corp. Be concise and friendly.

Step 3 — Connect a channel

Choose a messaging channel and follow the guided OAuth flow. BotPhi handles webhook registration, message parsing, and response formatting automatically.

Step 4 — Deploy

Click "Deploy" or run the CLI command. BotPhi provisions an isolated instance and starts serving requests within seconds.

botphi deploy my-first-bot

# Output:
# ✓ Instance provisioned (ap-south-1)
# ✓ Model connected (gpt-4o)
# ✓ Channel active (slack)
# ✓ Bot is live at https://my-first-bot.botphi.app