Skip to main content

Overview

The Actions & Tools section is the powerhouse of your AI agent’s operational capabilities. Here you configure what your agent can do during and after calls - from transferring conversations to human agents or SIP systems, to making intelligent API calls, and setting up sophisticated callback protocols.
This section transforms your agent from a simple conversationalist into a fully integrated business automation tool that can take real actions and connect with your existing systems.

Call Transfer

Configure how your agent intelligently transfers calls to human agents, departments, or external phone systems when needed.

Transfer Configuration

1

Enable Transfer Capability

Toggle the “Call Transfer” option to activate the transfer functionality for your agent.
2

Choose Transfer Destination

Select between traditional phone numbers or SIP trunk endpoints based on your infrastructure.
3

Configure Transfer Logic

Define when and how your agent should initiate transfers using smart prompting.
4

Test Transfer Flow

Validate that transfers work correctly with your phone system before going live.

Transfer Types

Traditional Phone TransferPerfect for transferring to mobile phones, landlines, or traditional business numbers.
  • Format: Use international format (+1234567890)
  • Best for: Human agents, customer service departments, sales teams
  • Reliability: High compatibility with all phone systems
  • Cost: Standard telephony rates apply
Configuration Example:
Transfer Number: +1-555-123-4567
Department: Sales Team

Transfer Prompt Configuration

The transfer prompt is crucial - it teaches your AI agent exactly when to transfer calls. Here are professionally crafted templates:
Se usa para transferir la llamada con una persona en los casos que se indican a continuación:

- El usuario rechaza directamente la propuesta o servicio
- El usuario solicita expresamente hablar con una persona humana
- El usuario muestra desconfianza hacia el asistente virtual
- Se necesita información específica que no está disponible en el sistema
- El usuario requiere autorización o gestión de datos personales sensibles
- Situaciones que requieren negociación o decisiones complejas
- Quejas o reclamaciones que necesitan atención personalizada

Advanced Transfer Settings

Context Preservation: When transferring calls, the agent can provide context to the human operator:
  • Customer’s name and contact information
  • Reason for the call and main topics discussed
  • Specific questions or requests that triggered the transfer
  • Customer sentiment and interaction history
Warm vs Cold Transfer:
  • Warm Transfer: Agent briefly speaks with the human before connecting the customer
  • Cold Transfer: Direct transfer without agent introduction
  • Conference: Three-way call with agent, customer, and human

End Call Strategy

Configure intelligent call termination with sophisticated answering machine detection and strategic conversation ending.

Call Termination Logic

1

Enable Smart Termination

Activate the “End Call Strategy” to handle call endings intelligently.
2

Configure Detection Phrases

Set up language-specific phrases that indicate answering machines or voicemail.
3

Define Completion Criteria

Specify what constitutes a successful call that should end naturally.
4

Handle Edge Cases

Configure responses for unexpected scenarios like long silences or system errors.

Answering Machine Detection

Our advanced detection system recognizes voicemail and answering machines across multiple languages:
Detection Accuracy: 95%+ accuracy rate in identifying automated systems Languages Supported: Spanish, English, French, German, Italian, Portuguese Response Time: Detection typically occurs within 3-5 seconds of answering
Common Detection Phrases:
  • “En este momento no podemos atenderle”
  • “Deje su mensaje después de la señal”
  • “Gracias por llamar, en estos momentos…”
  • “Bienvenido al buzón de voz de…”
  • “Este es el contestador de…”

Strategic Call Ending

Beyond answering machine detection, configure professional call termination for various scenarios: Natural Completion Triggers:
  • Customer explicitly says goodbye or indicates call completion
  • Agent has successfully completed the stated objective
  • All customer questions have been answered satisfactorily
  • Meeting or appointment has been successfully scheduled
Professional Termination Scenarios:
  • Customer declines service and requests no further contact
  • Call duration exceeds maximum configured time
  • Customer becomes unresponsive for extended period
  • System detects inappropriate or abusive language

Webhook Tools

Create powerful API integrations that allow your agent to interact with external systems during conversations.

Webhook Fundamentals

Real-time Integration: Execute API calls during live conversations Dynamic Parameters: Use conversation data to populate API requests
Response Processing: Agent can react to API responses and continue conversation Error Handling: Graceful fallbacks when APIs are unavailable

Creating Webhook Tools

1

Add New Integration

Click “Add Webhook Tool” to create a new API integration point.
2

Configure Tool Identity

  • Name: Clear, descriptive name (e.g., “create_lead”, “check_availability”)
  • Description: Explain when the AI should use this tool
  • Trigger Conditions: Define conversation scenarios that should activate the tool
3

Set API Endpoint

  • URL: The complete webhook endpoint URL
  • Method: Choose GET, POST, PUT, PATCH, or DELETE
  • Authentication: Configure headers for API security
4

Define Parameters

Map conversation data to API parameters using dynamic extraction.

Parameter Configuration

Each webhook parameter supports sophisticated data mapping:
PropertyDescriptionOptionsExample
NameParameter identifierCustom stringcustomer_name
TypeData validationstring, number, boolean, integer, array, objectstring
RequiredMandatory fieldtrue/falsetrue
DescriptionAI understandingNatural language”Customer’s full name from conversation”
Value SourceHow to get valueLLM extraction, static value, conversation metadataLLM prompt
Extraction PromptAI instructionNatural language”Extract the customer’s full name”

Advanced Parameter Types

URL ParametersAdded to the URL query string for GET requests or as additional parameters.
GET https://api.example.com/customers?name=John&phone=+1234567890&source=phone_call
Best for:
  • Search queries
  • Filtering parameters
  • API keys and tokens
  • Simple data passing

Webhook Execution Settings

Configure how and when webhooks execute: Execution Timing:
  • Immediate: Execute during the active call
  • Deferred: Execute after the call ends
  • Conditional: Execute based on conversation outcome
Agent Behavior:
  • Interruption Control: Should agent pause while waiting for API response?
  • Pre-tool Speech: Force agent to announce action before executing
  • Tool Call Sound: Play typing sound during execution
  • Response Timeout: Maximum wait time for API response (5-60 seconds)
Error Handling:
  • Retry Logic: Automatic retries on failure
  • Fallback Messages: What to say if API fails
  • Graceful Degradation: Continue conversation without API data

Practical Webhook Examples

Lead Creation Webhook
{
  "name": "create_crm_lead",
  "description": "Create a new lead in Salesforce when customer shows purchase interest",
  "url": "https://api.salesforce.com/services/data/v52.0/sobjects/Lead",
  "method": "POST",
  "headers": {
    "Authorization": "Bearer {{salesforce_token}}",
    "Content-Type": "application/json"
  },
  "body_parameters": {
    "FirstName": {
      "type": "string",
      "required": true,
      "description": "Customer's first name"
    },
    "LastName": {
      "type": "string", 
      "required": true,
      "description": "Customer's last name"
    },
    "Phone": {
      "type": "string",
      "required": true,
      "description": "Customer's phone number"
    },
    "Company": {
      "type": "string",
      "description": "Customer's company name if mentioned"
    },
    "LeadSource": {
      "type": "string",
      "static_value": "AI Phone Call"
    }
  }
}

Authentication Methods

{
  "Authorization": "Bearer your-jwt-token-here"
}
Common for: REST APIs, OAuth 2.0, JWT authentication

Recall Protocol

Set up sophisticated automatic callback systems for comprehensive customer follow-up when initial calls don’t achieve their objectives.

Recall System Architecture

Smart Scheduling: Intelligent timing based on customer behavior and business rules Success Detection: Automatic recognition of completed objectives to prevent unnecessary callbacks
Compliance Built-in: Respects Do Not Call lists and regulatory requirements Escalation Paths: Gradually increase intervals and change strategies over time

Recall Configuration

1

Enable Recall System

Activate the recall protocol and set maximum callback attempts.
2

Define Success Conditions

Specify what constitutes a successful call that doesn’t require follow-up.
3

Configure Callback Schedule

Set intelligent timing intervals between recall attempts.
4

Set Escalation Rules

Define how callback strategy changes with each attempt.

Recall Timing Strategies

AttemptTimingUse CaseSuccess Rate
1st Recall15-30 minutesBusy signal, no answer35%
2nd Recall2-4 hoursSame day follow-up25%
3rd RecallNext business dayProfessional follow-up20%
4th Recall3-5 days laterPersistent but respectful15%
5th Recall1-2 weeks laterFinal attempt10%

Success Condition Examples

Objective Achievement:
  • Customer provided required information (contact details, preferences)
  • Meeting or appointment successfully scheduled
  • Purchase decision made (positive or negative with clear finality)
  • Information successfully delivered (pricing, product details, etc.)
Explicit Customer Requests:
  • Customer explicitly requested no further contact
  • Customer scheduled their own callback time
  • Customer was transferred to human agent successfully
  • Customer asked to be contacted via different method (email, text)
Call Completion Scenarios:
  • Full conversation completed with all questions answered
  • Customer showed clear lack of interest with polite decline
  • Answering machine message successfully delivered
  • Technical issues prevented meaningful conversation

Advanced Recall Features

Dynamic Scheduling:
{
  "recall_attempts": [
    {
      "attempt": 1,
      "delay_minutes": 30,
      "strategy": "same_agent_voice",
      "message_variation": "brief_follow_up"
    },
    {
      "attempt": 2, 
      "delay_hours": 4,
      "strategy": "same_agent_voice",
      "message_variation": "value_proposition"
    },
    {
      "attempt": 3,
      "delay_days": 1,
      "strategy": "different_approach",
      "message_variation": "limited_time_offer"
    }
  ]
}
Callback Context Preservation:
  • Previous conversation summaries
  • Customer preferences and objections
  • Best times to call based on answer patterns
  • Failed attempt reasons and adjustments needed
Compliance Features:
  • Automatic DNC (Do Not Call) list checking
  • Time zone awareness for appropriate calling hours
  • Frequency limits (max calls per day/week)
  • Opt-out recognition and processing
Legal Compliance Required: Recall protocols must comply with local telecommunications laws like TCPA (US), GDPR (EU), and DNC regulations. Always include opt-out mechanisms and respect customer preferences.

Caller ID Configuration

Set up professional caller IDs for outbound calls to improve answer rates and brand recognition.

Caller ID Management

1

Add Phone Numbers

Register your business phone numbers for use as caller IDs.
2

Verify Ownership

Complete verification process via SMS or voice call.
3

Set Default Caller ID

Choose which number appears for outbound calls.
4

Monitor Performance

Track answer rates by caller ID to optimize selection.

Caller ID Best Practices

Number Selection:
  • Use local area codes when possible (increases answer rates by 20-30%)
  • Choose memorable, professional-looking numbers
  • Avoid numbers that look like spam (avoid 555, repeated digits)
  • Use toll-free numbers for national campaigns
Verification Requirements:
  • SMS verification for mobile numbers
  • Voice verification for landlines
  • Business documentation for toll-free numbers
  • Compliance with carrier requirements
Performance Optimization:
  • A/B test different caller IDs
  • Monitor answer rates by time of day
  • Adjust caller ID based on campaign type
  • Use different numbers for different business units

Best Practices

Call Transfer Optimization

  1. Clear Trigger Definition
    • Define specific, unambiguous scenarios for transfers
    • Avoid vague conditions that could cause unnecessary transfers
    • Include escalation paths for different types of issues
  2. Context Preservation
    • Ensure human agents receive comprehensive call context
    • Include customer information, conversation summary, and specific needs
    • Provide clear reason for transfer to minimize customer frustration
  3. Fallback Planning
    • Configure backup numbers for failed transfers
    • Set up alternative routing for business hours vs. after hours
    • Have contingency plans for system outages

Webhook Development

  1. API Design Principles
    • Use descriptive, semantic naming for webhook tools
    • Ensure APIs return meaningful success/error responses
    • Implement proper HTTP status codes and error messages
    • Design for idempotency to handle retries safely
  2. Security Implementation
    • Always use HTTPS endpoints
    • Implement proper authentication (Bearer tokens, API keys)
    • Validate and sanitize all incoming data
    • Use rate limiting to prevent abuse
  3. Performance Optimization
    • Keep API response times under 5 seconds when possible
    • Implement proper caching strategies
    • Use asynchronous processing for long-running operations
    • Monitor webhook performance and success rates

End Call Strategy

  1. Cultural Adaptation
    • Adapt detection phrases to local language variants
    • Respect cultural norms for call termination
    • Use appropriate farewell messages for different regions
    • Consider local business customs and communication styles
  2. Continuous Improvement
    • Monitor call completion rates and reasons
    • Analyze answering machine detection accuracy
    • Adjust prompts based on false positives/negatives
    • Regular review of termination criteria effectiveness

Recall Protocol Management

  1. Compliance First
    • Implement robust Do Not Call list checking
    • Respect time zone restrictions and business hours
    • Provide clear opt-out mechanisms in every interaction
    • Document consent and preferences carefully
  2. Strategic Timing
    • Use data-driven approach to optimize callback timing
    • Consider customer behavior patterns and preferences
    • Avoid over-communication that could damage relationships
    • Balance persistence with respect for customer wishes
  3. Success Measurement
    • Track callback success rates by attempt number
    • Measure customer satisfaction with follow-up processes
    • Monitor complaint rates and opt-out requests
    • Optimize based on actual conversion data

Troubleshooting Common Issues

Failed Transfers:
  • Check phone number format (use international format)
  • Verify SIP URI syntax and network connectivity
  • Test transfer destinations manually
  • Review carrier compatibility and routing
Context Loss:
  • Ensure transfer prompts include context instructions
  • Verify human agent training on receiving transferred calls
  • Check integration between systems
  • Review call logging and recording setup

Integration Examples

Complete Sales Follow-up Workflow

{
  "workflow_name": "comprehensive_sales_follow_up",
  "components": {
    "initial_call": {
      "greeting": "Dynamic with customer name",
      "objectives": ["present_offer", "gather_interest", "schedule_demo"]
    },
    "webhook_integrations": [
      {
        "name": "create_lead",
        "trigger": "customer_shows_interest",
        "crm_system": "salesforce"
      },
      {
        "name": "schedule_demo", 
        "trigger": "demo_requested",
        "calendar_system": "calendly"
      }
    ],
    "transfer_conditions": [
      "pricing_negotiation_needed",
      "technical_questions_beyond_scope",
      "customer_explicitly_requests_human"
    ],
    "recall_protocol": {
      "max_attempts": 3,
      "schedule": ["1_hour", "1_day", "1_week"],
      "success_conditions": ["demo_scheduled", "clear_rejection", "competitor_chosen"]
    }
  }
}

Coming Soon

Our development roadmap includes exciting new capabilities to further enhance your agent’s operational power:
FeatureDescriptionTimelineImpact
📧 Email IntegrationSend personalized emails during or after callsQ2 2024Medium
💬 SMS IntegrationText message capabilities for follow-upQ2 2024High
📱 WhatsApp IntegrationWhatsApp Business API for messagingQ3 2024High
🔄 Multi-step WorkflowsComplex, branching automation sequencesQ3 2024Very High
📊 Advanced AnalyticsDetailed performance metrics and optimization insightsQ3 2024Medium
🤖 AI-Powered OptimizationAutomatic improvement of webhook and transfer performanceQ4 2024Very High

Next Steps

Input Parameters

Configure dynamic variables and data inputs for personalized conversations

Output Settings

Set up data extraction and webhook delivery for comprehensive automation