AI Flows - Multi-Component AI Integration Platform

A comprehensive platform that integrates various AI-powered components for knowledge management and customer support operations.

Components

1. PTR Knowledge Base Assistant

A Python-based knowledge base assistant that integrates with OpenAI and AstraDB to provide intelligent responses to PTR-related queries.

Features

• Chat-based interface for knowledge base queries
• OpenAI-powered information extraction and response generation
• AstraDB integration for knowledge storage and retrieval
• Content summarization capabilities
• PTR-specific query filtering
• Comprehensive test suite

Usage

from components.ptr_kb_assistant import PTRKBAssistant
from dotenv import load_dotenv
import os

# Load environment variables
load_dotenv()

# Initialize the assistant
assistant = PTRKBAssistant(
    openai_api_key=os.getenv('OPENAI_API_KEY'),
    astra_db_config={
        "secure_connect_bundle": os.getenv('ASTRA_DB_SECURE_BUNDLE_PATH'),
        "client_id": os.getenv('ASTRA_DB_CLIENT_ID'),
        "client_secret": os.getenv('ASTRA_DB_CLIENT_SECRET'),
        "keyspace": os.getenv('ASTRA_DB_KEYSPACE')
    }
)

# Process a query
result = assistant.process_query("What are the key features of PTR's 5G solutions?")
print(result)

2. Outreach API Mock Server

A mock implementation of the Outreach API for testing and development purposes.

Features

• Full CRUD operations for leads, accounts, opportunities, sequences, and templates
• In-memory data storage with automatic ID generation
• Relationship management between resources
• Input validation and error handling
• Support for filtering and pagination

Running the Mock Server

uvicorn mockers.outreach.mock_outreach_api:app --reload --port 8000

3. Email Formation Component

Creates personalized emails using knowledge from various sources including Google Drive, file uploads, and LlamaCloud knowledge base.

Features

• LlamaCloud Knowledge Base integration
• Google Drive integration
• File upload support
• OpenAI-powered email generation
• Robust file handling

Required Fields for LlamaCloud

• LlamaCloud Index Name
• LlamaCloud Project Name
• LlamaCloud Organization ID
• LlamaCloud API Key

4. Kayako MCP Server

A FastAPI server that provides access to Kayako support tickets using the Claude MCP protocol.

Features

• Full MCP protocol implementation
• Streaming response support
• Comprehensive ticket data retrieval
• Search functionality
• MAXIS case handling
• Web-based testing interface

Installation

1. Clone the repository:

git clone <repository-url>
cd ai-flows

2. Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

Configuration

Create a .env file in the root directory with the following variables:

# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key

# AstraDB Configuration
ASTRA_DB_SECURE_BUNDLE_PATH=path/to/secure-connect-bundle.zip
ASTRA_DB_CLIENT_ID=your_client_id
ASTRA_DB_CLIENT_SECRET=your_client_secret
ASTRA_DB_KEYSPACE=your_keyspace

# Outreach Configuration
OUTREACH_API_KEY=your_outreach_api_key

# Kayako Configuration
KAYAKO_BASE_URL=https://your-kayako-instance.kayako.com
KAYAKO_EMAIL=your-email@example.com
KAYAKO_PASSWORD=your-password

# Server Configuration
HOST=0.0.0.0
PORT=8000

# LlamaCloud Configuration
LLAMACLOUD_API_KEY=your_llamacloud_api_key
LLAMACLOUD_ORG_ID=your_llamacloud_org_id
LLAMACLOUD_PROJECT_NAME=your_project_name
LLAMACLOUD_INDEX_NAME=your_index_name

Testing

Run the test suite:

pytest tests/

For more detailed test output:

pytest tests/ -v --log-cli-level=INFO

API Documentation

Once the server is running, visit:

• http://localhost:8000/docs for Swagger UI
• http://localhost:8000/redoc for ReDoc documentation

Security Notes

• Store sensitive credentials in the .env file
• The .env file should never be committed to version control
• Consider implementing additional security measures in production

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

License

MIT License - see LICENSE file for details