AstroGPT — Telegram (Pseudo)Astrology Bot powered by LLMs ✨

AstroGPT is a learning-scale Telegram bot based LLM reasoning.

Behind the scenes the bot

1. Parses free-form text with an LLM (via LangChain) to recognize intent and extract dates, zodiac signs, etc.
2. Runs actions such as get_daily_forecast, update_user_data, or unsubscribe.
3. Generates replies with short forecasts, situational advice, or astrology jokes.
4. Stores state in Postgres so users can pick up the conversation later.

Tech stack

• Python
• Telegram Bot API (webhook mode)
• LangChain + OpenAI GPT-4o (can swap to any chat-completion model)
• Postgres + SQLAlchemy + Alembic
• Docker / docker-compose for local dev
• Deployment to GCP Cloud Run

Get Started

Telegram Bot Token Setup

To use this code, you need to obtain a Telegram Bot Token. Follow these steps to get your token:

1. Open Telegram and search for the BotFather.
2. Start a chat with the BotFather and use the /start command.
3. Use the /newbot command to create a new bot.
4. Follow the prompts to name your bot and choose a username for it.
5. Once the bot is created, the BotFather will provide you with a bot token. This token is required to authenticate your bot with the Telegram API.

Important: Keep your bot token secure and do not share it publicly.

Once you have your bot token, you can use it in the code to interact with the Telegram API.

Run database migrations

• Ensure you have Python and Alembic installed.
• Add DATABASE_URL to your terminal session or export it as an environment variable:

  export DATABASE_URL=<your_postgres_connection_string>

• Verify the connection by running:

  alembic current

  This should display the current migration state. If it fails, check your DATABASE_URL and ensure the Postgres server is running.
• Run the following command to apply migrations:

  alembic upgrade head

Run Locally

1. Set up environment variables:

   • Create a file named dev.env in the root directory.
   • Add the following variables to dev.env:

     DATABASE_URL=<your_postgres_connection_string>
     TG_BOT_TOKEN=<your_telegram_bot_token>
     OPEN_AI_KEY=<your_openai_api_key>
     

2. Run database migrations

3. Start the bot locally:

   • Use Docker Compose to start the services:

     docker-compose up

Deployment to Cloud Run

1. Set up environment variables:

   • Create a file named prod.cloudrun.env.yaml.
   • Add the following variables to prod.cloudrun.env.yaml:

     DATABASE_URL: <your_postgres_connection_string>
     TG_BOT_TOKEN: <your_telegram_bot_token>
     OPEN_AI_KEY: <your_openai_api_key>

2. Run database migrations

3. Deploy to Cloud Run:

   • Use the provided deployment script:

     ./deploy.sh

Why this repo?

• Human-like ChatBot conversation UI
• Use LLM-reasoning instead rigid /commands.
• Minimal business logic - new skills are just new “actions” plus a prompt tweak.

Pull requests and feedback are welcome. Have fun, and may your planetary transits be favourable! 🪐