Sibyl Telegram Interface

AWS Lambda function that handles Telegram bot messages and interfaces with the backend service.

1. Create a Telegram bot and get the bot token from @BotFather

2. Deploy the sibyl-core CloudFormation stack first

   • This stack exports the API endpoint that this service depends on
   • The stack name should follow the pattern sibyl-core-{env} where env is either dev or prod

3. Store the bot token in AWS Parameter Store:

aws ssm put-parameter \
    --name "/sibyl/telegram/bot-token" \
    --value "YOUR_BOT_TOKEN" \
    --type "SecureString" \
    --description "Telegram Bot Token for Sibyl"

This step is required before deploying the stack. If you deploy without setting up the parameter, the deployment will fail with a clear error message telling you how to fix it.

1. Create a virtual environment and install dependencies:

python -m venv venv
source venv/bin/activate  # On Windows: .\venv\Scripts\activate
pip install -r requirements.txt

2. Configure AWS Parameter Store:

• Store the Telegram bot token securely in AWS Parameter Store at /sibyl/telegram/bot-token
• Ensure the Lambda function has appropriate IAM permissions to access the parameter

Store the Telegram bot token in AWS Parameter Store:

Using AWS CLI:

aws ssm put-parameter \
    --name "/sibyl/telegram/bot-token" \
    --value "YOUR_BOT_TOKEN" \
    --type "SecureString" \
    --description "Telegram Bot Token for Sibyl"

Or using AWS Console:

1. Go to AWS Systems Manager

2. Navigate to Parameter Store

3. Click "Create parameter"

4. Set:

   • Name: /sibyl/telegram/bot-token
   • Type: SecureString
   • Value: Your Telegram bot token

5. Click "Create parameter"

6. Deploy to AWS Lambda:

• Create a new Lambda function
• Set Python 3.12 as the runtime
• Set lambda_handler.lambda_handler as the handler
• Configure environment variables if needed
• The function URL will be automatically configured by SAM

1. Install AWS SAM CLI following the official instructions

2. Build the SAM application:

sam build

3. Deploy the application:

# First-time deployment
sam deploy --guided

# Subsequent deployments
sam deploy

The webhook is automatically configured during the first deployment using a Custom Resource in CloudFormation. You don't need to run any additional commands.

To verify the webhook setup:

curl https://api.telegram.org/bot<YOUR_BOT_TOKEN>/getWebhookInfo

If you need to manually set up the webhook, you can use the provided script:

./setup_webhook.py --stack-name <your-stack-name>

• IP validation for Telegram webhook requests
• Message length validation
• Secure token storage using AWS Parameter Store
• Request validation using Pydantic models
• Comprehensive error handling and logging

To add the backend API client:

1. Generate the Python client from your OpenAPI specification
2. Add the generated client to the project
3. Update the lambda_handler function to use the client for backend communication

TODO: Add test cases for:

• IP validation
• Message validation
• Backend API integration
• Telegram API integration