On-chain AI Agent on Telegram

Introduction

By following this example, we can build a Telegram bot that interacts with the Cronos network, allowing users to query blockchain information and perform transactions.

Prerequisites

Having the following running and read:

• AI-Agent-Service - See Self-Hosting AI-Agent-Server to run your local AI agent;

• AI-Agent-Signer-App - See "Magic Link" Signer (Optional for on-chain transaction execution)

• Telegram Bot Token

• conda

Step 1 - Environment Variables Setup

Set up out environment variables, which includes: Explorer API keys, OpenAI API keys and telegram bot token by:

export CRONOS_ZKEVM_TESTNET_API=your_explorer_api_key_here
export OPENAI_API_KEY=your_openai_api_key_here
export TELEGRAM_BOT_TOKEN=your_telegram_bot_token_here

Step 2 - Set Up Telegram Bot

1. Clone the repository and navigate to the bot directory:

git clone https://github.com/crypto-com/developer-platform-sdk-examples.git
cd developer-platform-sdk-examples/bot/cryptocom-ai-telegram-bot

1. Create a Python virtual environment:

conda create -n telegram-bot python=3.10
conda activate telegram-bot

1. Install dependencies:

pip install -r requirements.txt

Step 3 - Run everything

Now we can start the bot by:

python bot.py

Once the bot is running, test it to ensure it’s working as expected:

1. Open Telegram and search for your bot using the bot name you set up with @BotFather.

2. Send a sample command or message, such as /start, to see if the bot responds.

---

Troubleshooting

If you encounter any issues:

• Check your API keys to ensure they are correctly set in the environment variables.

• Verify dependencies by confirming that all packages in requirements.txt are installed.

• Review Telegram Bot API limitations to ensure compliance with rate limits and usage guidelines.