spacestep-tool-call 0.1.0

SpaceStep Tool Call

SpaceStep Tool Call

AI-powered call center automation and appointment scheduling library. Designed specifically for use with AI-based voice agents.

Features

• Call management (transfer, end call)
• Time slot availability checking
• Appointment booking
• Date and time utilities

Installation

# Installation from local directory
pip install -e ./spacestep_tool_call

# Or installation from PyPI (when published)
# pip install spacestep-tool-call

Basic Usage

import asyncio
from spacestep_tool_call import book_appointment, get_available_time_slots

# Check available time slots
async def check_slots():
    dates = ["2023-06-01", "2023-06-02", "2023-06-03"]
    slots = await get_available_time_slots(dates)
    print(slots)

# Book an appointment
async def make_booking():
    result = await book_appointment(
        name="John Smith",
        email="[e-mail]",
        phone_number="[phone]",
        selected_time_slot="2023-06-01, 14:00 - 14:30"
    )
    print(result)

asyncio.run(check_slots())
asyncio.run(make_booking())

Integration with Existing AI Voice Agent

Method 1: Direct Import Replacement (Minimal Changes)

This method allows you to use the library with minimal changes to existing code, simply by changing imports.

# Before:
from voice_agent.agent_zero.MainAgent.tools import (
    transfer_call, get_available_time_slots, book_appointment, 
    get_weekday, end_call, FUNCTION_CALLING_TOOLS
)

# After:
from spacestep_tool_call import (
    transfer_call, get_available_time_slots, book_appointment, 
    get_weekday, end_call, FUNCTION_CALLING_TOOLS
)

# The rest of the code remains unchanged

Method 2: Adapter for Voice Agent

This method uses an adapter that preserves the old interaction interface:

# voice_agent/agent_zero/MainAgent/tools.py

from spacestep_tool_call import (
    transfer_call as lib_transfer_call,
    get_available_time_slots as lib_get_available_time_slots,
    book_appointment as lib_book_appointment,
    get_weekday as lib_get_weekday,
    end_call as lib_end_call,
    FUNCTION_CALLING_TOOLS
)

# For full compatibility, we maintain the same function signatures
async def transfer_call():
    return await lib_transfer_call()

async def get_weekday(date: str):
    return await lib_get_weekday(date)

async def end_call():
    return await lib_end_call()

async def get_available_time_slots(dates):
    # Use environment variable as in the original code
    return await lib_get_available_time_slots(dates)

async def book_appointment(name, email, phone_number, selected_time_slot):
    # Use environment variable as in the original code
    return await lib_book_appointment(name, email, phone_number, selected_time_slot)

# FUNCTION_CALLING_TOOLS is already imported from the library above

Method 3: Direct Integration into Voice Agent

For maximum flexibility, you can directly integrate the library into the voice agent's code:

# voice_agent/agent_zero/MainAgent/agent.py
from spacestep_tool_call import (
    get_available_time_slots, book_appointment, 
    transfer_call, end_call, get_weekday, FUNCTION_CALLING_TOOLS
)

class VoiceAgent:
    def __init__(self):
        # Agent initialization
        self.tools = FUNCTION_CALLING_TOOLS
        
    async def handle_call(self, query):
        # Call handling logic
        # ...
        
        if "schedule appointment" in query:
            dates = ["2023-06-01", "2023-06-02", "2023-06-03"]
            available_slots = await get_available_time_slots(dates)
            # Further logic...
            
        elif "book slot" in query:
            result = await book_appointment(
                name="Client Name",
                email="[e-mail]",
                phone_number=query.get("phone", ""),
                selected_time_slot=query.get("slot", "")
            )
            # Process result...
            
        elif "transfer" in query:
            await transfer_call()
            
        elif "goodbye" in query:
            await end_call()

Configuring Webhook URLs

The library supports two methods for configuring webhooks:

1. Through environment variables (as in the original code):

import os

# Set environment variables
os.environ["TOOL_CHECK_SLOT_AVAILABILITY_WEBHOOK_URL"] = "https://example.com/check-availability"
os.environ["TOOL_BOOKING_WEBHOOK_URL"] = "https://example.com/book-appointment"

# Now functions will use these URLs
slots = await get_available_time_slots(dates)

2. By directly passing URLs to functions:

# Direct URL passing
slots = await get_available_time_slots(
    dates=["2023-06-01", "2023-06-02"],
    webhook_url="https://example.com/check-availability"
)

# Direct URL passing for booking
result = await book_appointment(
    name="Client",
    email="[e-mail]",
    phone_number="[phone]",
    selected_time_slot="2023-06-01, 14:00 - 14:30",
    webhook_url="https://example.com/book-appointment"
)

License

MIT