callbotics-sdk 1.0.0

Official Python SDK for Callbotics Voice AI Platform

Callbotics Python SDK

Official Python SDK for the Callbotics Voice AI Platform. Build, deploy, and manage voice AI bots with ease.

Features

• 🚀 Easy Bot Creation - Create complete voice bots in just a few lines of code
• 🔧 Full API Coverage - Access all Callbotics platform features
• 🎯 Type-Safe - Built with type hints for better IDE support
• 📦 Simple Installation - Install via pip
• 🎨 Flexible Configuration - Fine-tune every aspect of your voice agents
• 🔄 Campaign Management - Orchestrate large-scale calling campaigns
• 📊 Call Analytics - Track and analyze call performance

Installation

pip install callbotics-sdk

Or install from source:

git clone https://github.com/AM-Infoweb/callbotics-core-api.git
cd callbotics-core-api/callbotics-sdk
pip install -e .

Quick Start

1. Initialize the Client

from callbotics_sdk import CallboticsClient

# Initialize with your API credentials
client = CallboticsClient("https://api.callbotics.ai")
client.login("[e-mail]", "your-password")

# Or use an existing token
client = CallboticsClient(
    "https://api.callbotics.ai",
    api_token="your-jwt-token"
)

2. Create a Complete Bot (Easy Way)

The easiest way to create a bot is using the create_complete_bot() method:

bot = client.create_complete_bot(
    name="Customer Support Bot",

    # LLM Configuration
    llm_type="agent_chat_gpt",
    llm_model="gpt-4",
    llm_api_key="sk-...",
    llm_temperature=0.7,

    # Prompt Configuration
    prompt_background="You are a friendly customer support agent for Acme Corp.",
    prompt_business_logic="Always verify customer account before proceeding. Be helpful and professional.",
    prompt_conversational_tips="Use a warm, empathetic tone. Keep responses concise.",

    # Voice Configuration (Rime AI)
    voice_api_key="your-rime-key",
    voice_speaker="wildflower",
    voice_speed=1.2,

    # Transcriber Configuration (Deepgram)
    transcriber_api_key="your-deepgram-key",
    transcriber_language="en-US",

    # Telephony Configuration (Telnyx)
    telephony_provider="telnyx",
    telephony_auth_token="your-telnyx-token",
    telephony_connection_id="your-connection-id"
)

print(f"Bot created with ID: {bot['agent']['id']}")

3. Make a Call

call = client.calls.create(
    agent_id=bot['agent']['id'],
    to_number="[phone]",
    from_number="[phone]"
)

print(f"Call initiated: {call['id']}")

4. Create a Campaign

campaign = client.campaigns.create(
    name="Q1 Customer Outreach",
    agent_id=bot['agent']['id'],
    direction="outbound",
    concurrency=5,  # Max 5 simultaneous calls
    from_phone="[phone]",
    recording=True,
    initial_message="Hi, this is Sarah from Acme Corp."
)

# Start the campaign with a contact list
client.campaigns.start(
    campaign_id=campaign['id'],
    contact_list_id="your-contact-list-id"
)

Advanced Usage

Manual Bot Configuration

For more control, you can create each configuration separately:

# 1. Create LLM Configuration
llm_config = client.llm_configs.create(
    name="GPT-4 Config",
    llm_type="agent_chat_gpt",
    model="gpt-4",
    temperature=0.7,
    api_key="sk-..."
)

# 2. Create Prompt Configuration
prompt_config = client.prompt_configs.create(
    name="Support Prompt",
    background="You are a customer support agent",
    business_logic="Verify account before proceeding",
    conversational_tips="Be warm and professional"
)

# 3. Create Voice Configuration (Rime)
voice_config = client.voice_configs.create_rime(
    name="Rime Voice",
    api_key="your-rime-key",
    speaker="wildflower",
    model_id="mistv2",
    speed_alpha=1.2,
    reduce_latency=True
)

# 4. Create Transcriber Configuration (Deepgram)
transcriber_config = client.transcriber_configs.create_deepgram(
    name="Deepgram STT",
    api_key="your-deepgram-key",
    language="en-US",
    model="nova-2-phonecall"
)

# 5. Create Telephony Configuration (Telnyx)
telephony_config = client.telephony_configs.create_telnyx(
    name="Telnyx Config",
    auth_token="your-token",
    connection_id="your-connection-id"
)

# 6. Create Agent with all configs
agent = client.agents.create(
    name="My Custom Bot",
    llm_config_id=llm_config['id'],
    prompt_config_id=prompt_config['id'],
    voice_config_id=voice_config['id'],
    transcriber_config_id=transcriber_config['id'],
    telephony_config_id=telephony_config['id']
)

Managing Campaigns

# List all campaigns
campaigns = client.campaigns.list(page=1, per_page=10)

# Get specific campaign
campaign = client.campaigns.get(campaign_id="...")

# Update campaign
client.campaigns.update(
    campaign_id="...",
    concurrency=10,
    recording=True
)

# Pause/Resume campaign
client.campaigns.pause(campaign_id="...")

# Check available concurrency
concurrency_info = client.campaigns.get_available_concurrency()
print(f"Available slots: {concurrency_info}")

Managing Calls

# List all calls
calls = client.calls.list(
    page=1,
    per_page=20,
    status="ended",
    campaign_id="..."
)

# Get call details
call = client.calls.get(call_id="...")
print(f"Duration: {call['duration']} seconds")
print(f"Transcript: {call['transcript']}")

# Get call analytics
summary = client.calls.get_summary(
    campaign_id="...",
    start_date="2024-01-01",
    end_date="2024-01-31"
)
print(f"Total minutes: {summary['total_minutes']}")

Listing and Updating Configurations

# List all voice configurations
voices = client.voice_configs.list()

# Get specific config
voice = client.voice_configs.get(config_id="...")

# Update prompt
client.prompt_configs.update(
    config_id="...",
    background="Updated persona",
    business_logic="New rules"
)

# Delete configuration
client.llm_configs.delete(config_id="...")

Configuration Options

Supported LLM Providers

• OpenAI ChatGPT (agent_chat_gpt)
  • Models: gpt-4, gpt-3.5-turbo, etc.
• Anthropic Claude (agent_anthropic)
  • Models: claude-3-opus-20240229, claude-3-sonnet-20240229, etc.
• Groq (agent_groq)
  • Models: llama-3.1-70b-versatile, etc.
• Azure OpenAI (agent_azure_open_ai)

Supported Voice Providers

• Rime AI (rime) - Low-latency, high-quality TTS
  • Speakers: wildflower, bloom, etc.
  • Models: mistv2, mist, v1
• ElevenLabs (eleven_labs) - Natural-sounding voices
• Deepgram (deepgram) - Fast TTS
• PlayHT (play_ht)

Supported Transcriber Providers

• Deepgram (deepgram) - Industry-leading STT
  • Models: nova-2-phonecall, nova-2, enhanced
  • Languages: 30+ supported

Supported Telephony Providers

• Telnyx (telnyx)
• Twilio (twilio)

Error Handling

from callbotics_sdk import (
    CallboticsException,
    AuthenticationError,
    APIError,
    ResourceNotFoundError,
    ValidationError
)

try:
    call = client.calls.create(
        agent_id="invalid-id",
        to_number="[phone]"
    )
except AuthenticationError:
    print("Authentication failed. Please check your credentials.")
except ValidationError as e:
    print(f"Validation error: {e}")
except ResourceNotFoundError:
    print("Agent not found.")
except APIError as e:
    print(f"API error: {e.status_code} - {e}")
except CallboticsException as e:
    print(f"General error: {e}")

Examples

See the /examples directory for complete working examples:

• basic_bot.py - Create and use a simple bot
• advanced_bot.py - Advanced configuration and campaign management
• campaign_management.py - Running large-scale campaigns

Authentication

The SDK supports two authentication methods:

1. Email/Password Login

client = CallboticsClient("https://api.callbotics.ai")
token = client.login("[e-mail]", "password")
print(f"Logged in with token: {token}")

2. Pre-authenticated Token

client = CallboticsClient(
    "https://api.callbotics.ai",
    api_token="your-jwt-token"
)

3. Set Token Later

client = CallboticsClient("https://api.callbotics.ai")
# ... later
client.set_token("your-jwt-token")

API Reference

Client

• CallboticsClient(base_url, api_token=None) - Initialize client
• login(email, password) - Authenticate with credentials
• set_token(token) - Set authentication token manually
• create_complete_bot(...) - Create a bot with all configurations

Resources

• client.agents - Agent management
• client.campaigns - Campaign management
• client.calls - Call management
• client.llm_configs - LLM configuration
• client.voice_configs - Voice configuration
• client.transcriber_configs - Transcriber configuration
• client.telephony_configs - Telephony configuration
• client.prompt_configs - Prompt configuration

Requirements

• Python 3.8+
• requests>=2.28.0
• pydantic>=2.0.0

Support

• Documentation: https://docs.callbotics.com
• Issues: https://github.com/AM-Infoweb/callbotics-core-api/issues
• Email: [e-mail]

License

MIT License - see LICENSE file for details

Contributing

Contributions are welcome! Please submit pull requests or open issues on GitHub.

Changelog

v1.0.0 (2024)

• Initial release
• Full API coverage for agents, campaigns, and calls
• Configuration builders for all providers
• Comprehensive error handling
• Easy bot creation with create_complete_bot()