BotOS
Turn your AI agent into a messaging bot — step by step
Your AI agent can live inside messaging apps. Users chat with it like a friend — no app to install, no website to visit.
What Is a BotOS?
BotOS is your AI agent wrapped in a messaging app interface. Instead of typing prompts in a terminal, users send normal messages on Telegram, Discord, Slack, or WhatsApp.
| Concept | What It Means |
|---|---|
| Agent | The AI brain — understands questions, uses tools, generates answers |
| Bot | The messenger — receives messages from the app, sends them to the agent, delivers replies |
| Platform | Where users chat — Telegram, Discord, Slack, or WhatsApp |
Step-by-Step: How a Message Flows
| Step | What Happens |
|---|---|
| 1 | User types a message in their favorite app |
| 2 | The messaging platform sends it to your bot |
| 3 | Your bot passes the message to the AI agent |
| 4 | The agent thinks, searches the web, checks documents — whatever you’ve configured |
| 5 | The agent sends an answer back through the bot |
| 6 | The user sees the reply in their chat |
Which Platform Should I Use?
| Platform | Best For | Setup Difficulty |
|---|---|---|
| Telegram | Personal bots, public bots, quick prototypes | ⭐ Easy |
| Discord | Communities, gaming, developer groups | ⭐ Easy |
| Slack | Workplaces, teams, internal tools | ⭐⭐ Medium |
| Business communication, customer support | ⭐⭐⭐ Advanced |
Getting Started — 3 Ways
Pick the way that fits you best:
CLI (Easiest)
No code needed. One command:
# Telegram
praisonai bot telegram --token $TELEGRAM_BOT_TOKEN
# Discord
praisonai bot discord --token $DISCORD_BOT_TOKEN
# Slack
praisonai bot slack --token $SLACK_BOT_TOKEN --app-token $SLACK_APP_TOKENA default AI assistant is created automatically.
Python
from praisonaiagents import Agent
from praisonai.bots import TelegramBot
agent = Agent(
name="assistant",
instructions="You are a helpful assistant",
llm="gpt-4o-mini"
)
bot = TelegramBot(token="YOUR_BOT_TOKEN", agent=agent)
import asyncio
asyncio.run(bot.start())YAML
platform: telegram
token: "${TELEGRAM_BOT_TOKEN}"
agent:
name: "My Assistant"
instructions: "You are a helpful AI assistant."
llm: "gpt-4o-mini"praisonai bot start --config bot.yamlAdding Superpowers
Your bot can do more than just chat. Add capabilities with simple flags:
| Capability | CLI Flag | What It Does |
|---|---|---|
| Memory | --memory | Remembers what users said in previous messages. Safe for multi-worker deployments — messages survive concurrent updates |
| Knowledge | --knowledge | Answers from your PDF, text, or markdown files |
| Web Search | --web | Searches the internet for up-to-date info |
| Thinking | --thinking high | Takes more time to give better, deeper answers |
# A bot that remembers conversations and searches the web
praisonai bot telegram --token $TOKEN --memory --webBot vs Gateway
Two ways to deploy your agent for real-time chat:
| Bot | Gateway | |
|---|---|---|
| Platforms | One at a time | All at once |
| Agents | One per bot | Multiple, with routing |
| Best for | Quick start, single platform | Production, multi-channel |
| Command | praisonai bot telegram | praisonai gateway --config gateway.yaml |
Best Practices
Start with Telegram
Telegram is the easiest platform to set up — just message @BotFather and get a token in seconds.
Add memory for better conversations
Enable --memory so your bot remembers context between messages. Users get a much better experience.
Keep instructions simple
Write bot instructions like you’re talking to a person: “You are a helpful assistant for our customer support team. Be friendly and concise.”
Test locally, deploy with Docker
Use Socket Mode or polling for local testing. When ready for production, deploy with Docker and webhook mode.