Finance Banking Agent Template

A lightweight banking conversational agent template for the fictional Fenlo Bank that handles account management, card services, and money transfers.

This bot is from Hello Rasa.

Important This is configured to use OpenAI models. See config.yml

Use with Rasa-Pro 3.13

🚀 What's Included

This template provides a banking assistant with:

• Account Management: Balance checking and statement downloads
• Card Services: Card activation, blocking, replacement, and listing
• Money Transfers: Account-to-account transfers and third-party payments
• Contact Management: Add, list, and remove trusted contacts
• Bill Management: Bill payment reminders and scheduling
• Banking Knowledge: FAQ system with Fenlo Bank documentation

📁 Directory Structure

├── actions/         # Custom Python logic for banking operations
├── data/            # Banking conversation flows and training data
├── domain/          # Banking agent configuration
├── db/              # Mock JSON database for testing
├── docs/            # Fenlo Bank knowledge base and FAQ documents
└── config.yml       # Training pipeline configuration

Installation

Prerequisites:

• Rasa Pro license
• Python - supported Python versions: >=3.10.0,<3.14. For example, you can use pyenv to install Python 3.11.11: pyenv install 3.11.11

After you cloned the repository, follow these installation steps:

1. Navigate to the cloned repo:

   cd lena-hello-rasa

2. Set up the Python environment with pyenv or any other tool that manages Python versions:

   pyenv virtualenv 3.11.11 hello-rasa  # create a virtual environment
   pyenv activate hello-rasa             # activate your virtual environment
   pyenv local hello-rasa                # set auto-activation for this directory

3. Install the dependencies using uv:

   pip install uv
   uv pip install -r requirements.txt

4. Create an environment file .env in the root of the project with the following content:

   RASA_PRO_LICENSE=<your rasa pro license key>
   OPENAI_API_KEY=<your openai api key>

Useful commands

Train a model

rasa train --data data --domain domain

Talk to the bot

GUI-based interaction using Rasa inspector:

rasa inspect

Run E2E tests

rasa test e2e e2e_tests/

More about e2e tests: https://rasa.com/docs/reference/testing/test-cases