Basic Telegram Agent

Gemini agent with session persistence on Telegram

Code

1from kern.agent import Agent
2from kern.db.sqlite import SqliteDb
3from kern.models.google import Gemini
4from kern.os.app import AgentOS
5from kern.os.interfaces.telegram import Telegram
6
7agent_db = SqliteDb(session_table="telegram_sessions", db_file="tmp/telegram_basic.db")
8
9telegram_agent = Agent(
10    name="Telegram Bot",
11    model=Gemini(id="gemini-2.5-pro"),
12    db=agent_db,
13    instructions=[
14        "You are a helpful assistant on Telegram.",
15        "Keep responses concise and friendly.",
16        "When in a group, you respond only when mentioned with @.",
17    ],
18    add_history_to_context=True,
19    num_history_runs=3,
20    add_datetime_to_context=True,
21    markdown=True,
22)
23
24agent_os = AgentOS(
25    agents=[telegram_agent],
26    interfaces=[
27        Telegram(
28            agent=telegram_agent,
29            reply_to_mentions_only=True,
30        )
31    ],
32)
33app = agent_os.get_app()
34
35if __name__ == "__main__":
36    agent_os.serve(app="basic:app", reload=True)

Usage

Set up your virtual environment

1uv venv --python 3.12
2source .venv/bin/activate

1uv venv --python 3.12
2.venv\Scripts\activate

Set Environment Variables

1export TELEGRAM_TOKEN=your-bot-token-from-botfather
2export GOOGLE_API_KEY=your-google-api-key
3export APP_ENV=development

Install dependencies

1uv pip install -U "kern-ai[telegram]"

Run Example

1python basic.py

Key Features

Telegram Integration: Responds to direct messages and group @mentions
Conversation History: Maintains context with last 3 interactions
Persistent Memory: SQLite database for session storage
Group Chat Support: Only responds when mentioned in groups
DateTime Context: Time-aware responses