Bots

Understand what a bot is in BotPhi and how it maps to your AI assistant.

A Bot is the central resource in BotPhi. It represents a deployed AI assistant — including its model configuration, system prompt, connected channels, and runtime settings.

Bot Lifecycle

Created — Bot config is saved but not yet deployed
Deploying — Instance is being provisioned
Running — Bot is live and serving requests
Paused — Bot is stopped but retains its configuration
Failed — Deployment encountered an error

Bot Configuration

Every bot has a configuration object that defines its behavior. You can modify this from the dashboard or via the API.

bot-config.json

{
  "name": "support-bot",
  "model": {
    "provider": "anthropic",
    "model": "claude-sonnet-4-5-20250929",
    "temperature": 0.5,
    "max_tokens": 4096
  },
  "system_prompt": "You are a helpful support agent for Acme Corp.",
  "channels": ["slack", "web-widget"],
  "rate_limit": {
    "requests_per_minute": 60,
    "tokens_per_minute": 100000
  }
}

Multi-Model Support

BotPhi is model-agnostic. You can switch between providers without changing your bot's channels or system prompt. Supported providers:

OpenAI — GPT-4o, GPT-4o-mini, o1, o3-mini
Anthropic — Claude Opus 4, Claude Sonnet 4.5, Claude Haiku 4.5
Google — Gemini 2.0 Flash, Gemini 2.5 Pro
Meta — Llama 3.3, Llama 4 (via hosted providers)
Custom — Any OpenAI-compatible API endpoint

Installation

Channels