ShrutiAI 1.0.0

Python SDK for interacting with the shrutiAI API - your AI-powered assistant

shrutiAI SDK

A Python SDK for interacting with the shrutiAI API - your AI-powered assistant for chat, location services, OTP verification, and more.

Installation

pip install ShrutiAI

🔐 Authentication

All API requests require authentication. The SDK automatically handles this for you:

from ShrutiAI import ShrutiAIClient

client = ShrutiAIClient(api_key="your-secret-api-key")

Direct API access is not permitted. All requests must go through the authenticated SDK client.

Quick Start

from ShrutiAI import ShrutiAIClient

# Initialize the client
client = ShrutiAIClient(api_key="your-api-key")

# Send a chat message
response = client.chat(message="Hello, how are you?")
print(response['response'])

Features

• 🤖 AI Chat: Natural language conversations with AI assistant
• 📱 OTP Verification: Send OTP via AWS SNS, Twilio, or MSG91
• 📍 Location Services: Location-based queries and place searches
• 🖼️ Image Analysis: Analyze images with AI
• 📄 Document Processing: Extract and analyze text from documents
• 🌾 Farming Assistant: Specialized agricultural advice
• ✈️ Travel Information: Flight and train tracking
• 🔍 Web Search: Intelligent web search capabilities
• 📊 Math & Graphs: Mathematical problem solving and plotting
• 🎥 YouTube Integration: Find relevant video content

Authentication

You need an API key to use the shrutiAI SDK. Contact Shri Sai Technology to obtain your API key.

client = ShrutiAIClient(
    api_key="your-api-key",
    base_url="https://api.shruti.ai"  # Optional: defaults to localhost
)

Usage Examples

Basic Chat

from ShrutiAI import ShrutiAIClient

client = ShrutiAIClient(api_key="your-api-key")

# Everything goes through the single chat() method!
# AI automatically determines which tools to use based on your message

# Simple conversation
response = client.chat(message="Hello, how are you?")
print(response['response'])

# Location-based queries
response = client.chat(
    message="Find pizza restaurants near me",
    latitude="12.9716",
    longitude="77.5946",
    language="english"
)
print(response['response'])

# Any type of query - AI handles it all!
response = client.chat(message="What's the weather in Bangalore?")
response = client.chat(message="Get train status for Rajdhani Express")
response = client.chat(message="Solve this math problem: 2x + 3 = 7")
response = client.chat(message="Find hospitals nearby", latitude="12.9716", longitude="77.5946")

### OTP Verification

```python
# Send OTP via AWS SNS
result = client.verify_otp_aws("[phone]")
print(f"OTP sent: {result['otp']}")

# Send OTP via Twilio
result = client.verify_otp_twilio("[phone]")
print(f"OTP sent: {result['otp']}")

# Send OTP via MSG91
result = client.verify_otp_msg91("[phone]")
print(f"OTP sent: {result['otp']}")

Image Analysis

# Analyze an image
response = client.chat(
    message="What's in this image?",
    image_path="/path/to/your/image.jpg"
)
print(response['response'])

Document Processing

# Process a PDF document
response = client.chat(
    message="Summarize this document",
    file_path="/path/to/document.pdf"
)
print(response['response'])

Context Manager Usage

from ShrutiAI import ShrutiAIClient

# Use as context manager for automatic cleanup
with ShrutiAIClient(api_key="your-api-key") as client:
    response = client.chat(message="Hello!")
    print(response['response'])
    # Session automatically closed

Conversation Management

# Maintain conversation context
conversation_id = None

# First message
response = client.chat(
    message="Hi, I need help with farming",
    conversation_id=conversation_id
)
conversation_id = response.get('conversation_id')

# Follow-up message
response = client.chat(
    message="What crops should I grow in winter?",
    conversation_id=conversation_id
)
print(response['response'])

Error Handling

The SDK provides specific exception types for different error conditions:

from ShrutiAI import (
    ShrutiAIClient,
    AuthenticationError,
    RateLimitError,
    NotFoundError,
    ValidationError,
    ServerError,
    NetworkError
)

try:
    client = ShrutiAIClient(api_key="your-api-key")
    response = client.chat(message="Hello!")
except AuthenticationError as e:
    print(f"Authentication failed: {e}")
except RateLimitError as e:
    print(f"Rate limit exceeded: {e}")
except ValidationError as e:
    print(f"Invalid input: {e}")
except ServerError as e:
    print(f"Server error: {e}")
except NetworkError as e:
    print(f"Network error: {e}")
except Exception as e:
    print(f"Unexpected error: {e}")

Advanced Configuration

Custom Timeout

client = ShrutiAIClient(
    api_key="your-api-key",
    base_url="https://api.shruti.ai",
    timeout=60  # 60 seconds timeout
)

Multi-language Support

# Chat in different languages
response = client.chat(
    message="¿Cómo estás?",
    language="spanish"
)

response = client.chat(
    message="Comment allez-vous?",
    language="french"
)

response = client.chat(
    message="Wie geht es dir?",
    language="german"
)

API Methods

Core Methods

• chat() - Send messages to AI assistant
• verify_otp_aws() - Send OTP via AWS SNS
• verify_otp_twilio() - Send OTP via Twilio
• verify_otp_msg91() - Send OTP via MSG91
• health_check() - Check API health
• get_api_info() - Get API information

Helper Methods

The SDK provides a few helper methods for common use cases, but everything ultimately uses the single chat() method:

Helper Methods:

• ask_about_restaurants() - Helper for restaurant queries
• ask_about_hospitals() - Helper for hospital queries
• ask_about_weather() - Helper for weather queries

Single Entry Point Philosophy: All functionality goes through the chat() method - just like your backend's /api/chat-tool endpoint! The AI backend automatically determines which tools to use based on your message content. Simply ask questions naturally and the AI will handle the rest!

Example Usage:

# Everything uses the same chat() method
client.chat(message="Find pizza restaurants near me")
client.chat(message="What's the weather in Bangalore?")
client.chat(message="Get train status for Rajdhani Express")
client.chat(message="Solve: 2x + 3 = 7")
client.chat(message="What crops should I grow in winter?")

Parameters

Chat Method Parameters

• message (str): The message to send to the AI
• user_id (str, optional): User identifier for conversation tracking
• conversation_id (str, optional): Conversation identifier
• latitude (str, optional): Latitude for location-based queries
• longitude (str, optional): Longitude for location-based queries
• language (str): Response language (default: "english")
• image_path (str, optional): Path to image file for analysis
• file_path (str, optional): Path to document file for analysis

Response Format

All methods return a dictionary with the following structure:

{
    "response": "AI response text",
    "conversation_id": "conversation-identifier",
    "urls": {
        "restaurants": [...],
        "places": [...],
        "hospitals": [...],
        "pharmacies": [...],
        "groceries": [...],
        "accomodations": [...],
        "travel_info": {
            "trains": [...],
            "buses": [...],
            "cabs": [...],
            "flights": [...]
        },
        "flight_details": {...},
        "soil_details": {...},
        "math_graph_plotter": {...},
        "train_details": {...},
        "repair_service_result": [...],
        "items": [...],
        "extra_info": {...},
        "youtube_result": [...]
    }
}

Supported Languages

The SDK supports multiple languages for responses:

• English (default)
• Spanish
• French
• German
• Hindi
• And many more...

Rate Limits

The API has rate limits to ensure fair usage. The SDK automatically handles rate limit errors and raises RateLimitError exceptions when limits are exceeded.

Contributing

Contributions are welcome! Please see our Contributing Guide for details.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Support

• 📧 Email: [e-mail]
• 🌐 Website: https://shruti.ai/
• 🏢 Company: Shri Sai Technology (SST) India

Changelog

Version 1.0.0

• Initial release
• Basic chat functionality
• OTP verification (AWS, Twilio, MSG91)
• Image and document processing
• Location-based services
• Multi-language support
• Comprehensive error handling