AI Chatbot provides a Python framework for building sophisticated chatbots with minimal coding, targeting developers who need flexible, intelligent conversational agents. It offers local AI inference, seamless integration of custom Python functions and REST APIs, and dynamic learning capabilities, significantly reducing development effort and operational costs associated with external AI services.

How It Works

This project utilizes a hybrid conversational architecture, combining a declarative template engine with a powerful AI fallback mechanism. The template engine, using a custom syntax, manages conversational flows, state (memory, conditions, topics), and integrates external logic via Python functions or REST APIs. When a user query does not match a defined template, an Ollama-powered AI model (Llama 3.2) generates a response locally, ensuring privacy and eliminating API costs, thereby blending structured, predictable dialogue with flexible, generative AI capabilities.

Quick Start & Requirements

• Primary Install: pip install chatbotAI (includes Ollama setup prompt). Alternatively, install from source via GitHub clone and pip install -r requirement.txt, followed by python setup.py install.
• Prerequisites: Ollama must be installed for AI responses; specific setup instructions are in OLLAMA_SETUP.md. Python environment required.
• Demo: A simple demo function is available: from chatbot import demo; demo().
• Documentation: Sample code and integration guides for Facebook Messenger and Microsoft Bot Connector are linked within the README.

Highlighted Details

• Ollama & Llama 3.2 Integration: Enables state-of-the-art, local AI inference for human-like responses without API costs or internet dependency.
• Hybrid Conversational Model: Seamlessly blends template-based logic (memory, conditions, topic switching, recursion) with generative AI fallback for unmatched queries.
• Extensible Functionality: Supports deep integration of custom Python functions and complex REST API interactions directly within conversation templates.
• Dynamic Learning: Features online learning capabilities, allowing the bot to dynamically acquire new responses from interactions or explicit training.

Maintenance & Community

No specific details regarding maintainers, community channels (like Discord/Slack), or roadmap were provided in the README.

Licensing & Compatibility

The license type and compatibility notes for commercial use or closed-source linking are not specified in the provided README content.

Limitations & Caveats

The AI fallback functionality is dependent on a correctly configured Ollama installation and available models. If the bot is untrained and a query does not match any template, it will provide a default message indicating insufficient training data.