AI Assistant API Integration Guide
# AI Assistant API Integration Guide Transform your AI assistants into seamless extensions of your creator brand across any platform. Selgora's AI Assistant API enables you to embed intelligent, cont...
AI Assistant API Integration Guide
Transform your AI assistants into seamless extensions of your creator brand across any platform. Selgora's AI Assistant API enables you to embed intelligent, context-aware customer support directly into external GPTs, websites, and applications, creating cohesive customer experiences that maintain access control and usage tracking.
The Power of Distributed AI Assistance
Success scales when your expertise becomes accessible everywhere your customers need help. By integrating Selgora's AI Assistant API with external platforms like ChatGPT, you create omnipresent support that maintains subscriber verification, tracks usage analytics, and provides consistent value regardless of where customers interact with your AI.
Why External AI Integration Matters
Expanded Accessibility: Meet customers where they already spend time, whether that's ChatGPT, custom applications, or integrated business tools.
Seamless Experience: Customers access your AI expertise without switching platforms or remembering multiple login credentials.
Usage Analytics: Track AI assistant utilization across all integration points to understand customer behavior and optimize your AI training.
Revenue Protection: Maintain subscriber verification and access control even when AI interactions occur outside Selgora's platform.
Understanding Selgora's AI Assistant Architecture
Access Verification System
Real-time Subscription Checking: Every AI interaction triggers a verification request to Selgora, ensuring only active subscribers can access your AI assistance.
Automatic Access Management: When subscriptions expire or are cancelled, AI access is immediately revoked across all integration points without manual intervention.
Usage Tracking and Analytics: All AI interactions are logged with detailed metrics including:
- User identification and verification status
- Interaction timestamps and frequency
- Success/failure rates for access attempts
- Geographic and usage pattern analytics
API Key Authentication
Unique Assistant Keys: Each AI assistant generates a unique API key that identifies both the assistant and the creator tenant:
Key Format: sk_live_[40-character-string]
Example: sk_live_abc123def456ghi789jkl012mno345pqr678stu
Automatic Generation: API keys are created automatically when you create new AI assistants, ensuring immediate integration capability.
Setting Up Your AI Assistant for External Access
Creating AI Assistants in Selgora
Navigate to AI Assistant Management:
- Dashboard → AI Assistants → Create New Assistant
- Configure assistant name and description
- Set up knowledge base and response parameters
- Link assistant to specific offers for access control
- Copy the generated API key for external integration
Essential Configuration:
- Assistant Name: Clear, descriptive name for customer recognition
- Description: Purpose and capabilities explanation
- Linked Offers: Specific subscriptions that grant access
- Usage Limits: Optional rate limiting for cost control
GPT Integration Setup
Custom GPT Configuration: When creating custom GPTs that integrate with your Selgora AI assistant:
System Prompt Example:
You are [Assistant Name], a specialized assistant created by [Your Name] to help customers with [specific expertise area].
Before responding to any user request, you MUST verify their access by calling the Selgora verification API with their email address.
If verification succeeds, provide helpful, detailed responses within your area of expertise.
If verification fails, politely inform them they need an active subscription and provide the signup URL.
Always maintain a professional, helpful tone that reflects [Your Name]'s brand values.
API Action Configuration:
{
"openapi": "3.0.0",
"info": {
"title": "Selgora AI Assistant Access Verification",
"version": "1.0.0"
},
"servers": [
{
"url": "https://app.selgora.com/api"
}
],
"paths": {
"/v1/verify-access": {
"post": {
"summary": "Verify customer access to AI assistant",
"requestBody": {
"required": true,
"content": {
"application/json": {
"schema": {
"type": "object",
"properties": {
"user_email": {
"type": "string",
"description": "Customer's email address"
}
},
"required": ["user_email"]
}
}
}
},
"responses": {
"200": {
"description": "Access verification result",
"content": {
"application/json": {
"schema": {
"type": "object",
"properties": {
"access": {
"type": "boolean"
},
"message": {
"type": "string"
},
"signup_url": {
"type": "string"
}
}
}
}
}
}
},
"security": [
{
"BearerAuth": []
}
]
}
}
},
"components": {
"securitySchemes": {
"BearerAuth": {
"type": "http",
"scheme": "bearer"
}
}
}
}
Access Verification API Details
Endpoint Information
URL: https://app.selgora.com/api/v1/verify-access
Method: POST
Authentication: Bearer token using your AI assistant's API key
Rate Limit: 60 requests per minute
Request Format
Headers:
Authorization: Bearer sk_live_your_assistant_api_key
Content-Type: application/json
Request Body:
{
"user_email": "customer@example.com"
}
Alternative Parameter Name:
The API accepts either user_email
or email
as the parameter name for flexibility with different integration systems.
Response Formats
Successful Verification (Customer Has Access):
{
"access": true,
"message": "Active subscription"
}
Access Denied - User Not Found:
{
"access": false,
"message": "You don't have access to this assistant. Please sign up for an offer to get access.",
"signup_url": "https://your-landing-page-url.com"
}
Access Denied - Subscription Expired:
{
"access": false,
"message": "Your subscription has expired or you don't have an active offer. Please renew your subscription.",
"signup_url": "https://your-landing-page-url.com"
}
Authentication Error:
{
"message": "Unauthenticated."
}
Implementation Examples
GPT-4 Integration with Access Control
GPT System Behavior:
import requests
import json
def verify_customer_access(email, api_key):
"""Verify if customer has access to AI assistant"""
url = "https://app.selgora.com/api/v1/verify-access"
headers = {
"Authorization": f"Bearer {api_key}",
"Content-Type": "application/json"
}
data = {"user_email": email}
try:
response = requests.post(url, headers=headers, json=data)
return response.json()
except Exception as e:
return {"access": False, "message": "Verification failed"}
def handle_customer_request(customer_email, customer_question, api_key):
"""Handle customer request with access verification"""
# Verify access first
access_result = verify_customer_access(customer_email, api_key)
if access_result.get("access"):
# Customer has access, provide assistance
return provide_ai_assistance(customer_question)
else:
# Access denied, provide signup information
message = access_result.get("message", "Access verification failed")
signup_url = access_result.get("signup_url", "")
return f"{message}\n\nGet access here: {signup_url}"
def provide_ai_assistance(question):
"""Provide AI assistance for verified customers"""
# Your AI assistant logic here
return "Helpful AI response based on your expertise..."
Custom Web Application Integration
JavaScript Implementation:
class SelgoraAIAssistant {
constructor(apiKey) {
this.apiKey = apiKey;
this.baseUrl = 'https://app.selgora.com/api/v1';
}
async verifyAccess(email) {
try {
const response = await fetch(`${this.baseUrl}/verify-access`, {
method: 'POST',
headers: {
'Authorization': `Bearer ${this.apiKey}`,
'Content-Type': 'application/json'
},
body: JSON.stringify({ user_email: email })
});
return await response.json();
} catch (error) {
return {
access: false,
message: 'Verification service unavailable'
};
}
}
async handleChatMessage(userEmail, message) {
const accessCheck = await this.verifyAccess(userEmail);
if (accessCheck.access) {
// Provide AI assistance
return this.generateAIResponse(message);
} else {
// Return access denied message
return {
type: 'access_denied',
message: accessCheck.message,
signupUrl: accessCheck.signup_url
};
}
}
generateAIResponse(message) {
// Your AI response logic here
return {
type: 'ai_response',
message: 'AI-generated helpful response...'
};
}
}
// Usage
const assistant = new SelgoraAIAssistant('sk_live_your_api_key');
Mobile App Integration
React Native Example:
import AsyncStorage from '@react-native-async-storage/async-storage';
class SelgoraAIClient {
constructor(apiKey) {
this.apiKey = apiKey;
}
async verifyUserAccess() {
try {
const userEmail = await AsyncStorage.getItem('userEmail');
if (!userEmail) {
return { access: false, message: 'Please provide your email' };
}
const response = await fetch('https://app.selgora.com/api/v1/verify-access', {
method: 'POST',
headers: {
'Authorization': `Bearer ${this.apiKey}`,
'Content-Type': 'application/json'
},
body: JSON.stringify({ user_email: userEmail })
});
return await response.json();
} catch (error) {
return { access: false, message: 'Connection error' };
}
}
async sendMessage(message) {
const accessCheck = await this.verifyUserAccess();
if (accessCheck.access) {
// Show AI interface and provide assistance
return this.renderAIChat(message);
} else {
// Show subscription prompt
return this.renderSubscriptionPrompt(accessCheck);
}
}
}
Usage Analytics and Monitoring
Built-in Analytics Tracking
Automatic Logging: Every access verification request generates detailed logs including:
- Timestamp and user identification
- Access granted/denied status
- Reason for denial (no subscription, expired, etc.)
- Geographic location and user agent information
Analytics Dashboard: View AI assistant performance through Selgora's analytics:
- Total verification requests over time
- Success vs. failure rates
- Customer usage patterns and frequency
- Geographic distribution of usage
Custom Analytics Integration
Webhook Notifications: Configure webhooks to receive real-time notifications about AI assistant usage:
- Successful access verifications
- Access denial patterns
- High-usage customers requiring attention
- API key usage approaching limits
External Analytics Platforms: Integrate usage data with your existing analytics stack:
- Google Analytics for customer journey tracking
- Mixpanel for user behavior analysis
- Custom dashboards for business intelligence
Advanced Integration Patterns
Multi-Assistant Management
Assistant Specialization: Create multiple AI assistants for different customer segments or expertise areas:
- Beginner-focused assistant for new customers
- Advanced assistant for premium subscribers
- Industry-specific assistants for niche audiences
Progressive Access Unlocking: Use different assistants for different subscription tiers:
- Basic subscribers access foundational assistant
- Premium subscribers unlock advanced AI capabilities
- VIP subscribers gain access to specialized industry assistants
Cross-Platform Consistency
Brand Voice Maintenance: Ensure consistent AI personality and expertise across all integration points through:
- Standardized system prompts and training
- Regular assistant performance monitoring
- Customer feedback collection and optimization
- Consistent response quality standards
Knowledge Base Synchronization: Keep AI assistant knowledge current across all platforms:
- Regular training data updates
- Version control for assistant capabilities
- Consistent response accuracy standards
- Performance optimization based on usage analytics
Troubleshooting Common Issues
Access Verification Problems
API Key Issues:
- Verify API key format and validity
- Check that key corresponds to correct AI assistant
- Ensure key has necessary permissions
- Monitor key usage for rate limit breaches
Customer Email Problems:
- Validate email format before API calls
- Handle case sensitivity appropriately
- Account for customer email changes
- Provide clear error messages for invalid emails
Integration Performance
Response Time Optimization:
- Implement caching for frequent access checks
- Use asynchronous processing where possible
- Monitor API response times and optimize
- Implement fallback mechanisms for service unavailability
Rate Limit Management:
- Monitor API usage against limits
- Implement exponential backoff for retries
- Cache verification results when appropriate
- Consider usage patterns in system design
Success Indicators:
- AI assistant access verification working reliably
- Customer satisfaction with AI assistance quality
- Usage analytics showing consistent engagement
- Zero unauthorized access to AI capabilities
- Smooth integration with external platforms
Integration Best Practices:
- Always verify access before providing AI assistance
- Provide clear messaging for access denied scenarios
- Monitor usage patterns and optimize performance
- Maintain consistent brand voice across all platforms
- Implement proper error handling and fallback mechanisms
Remember: Your AI assistant API integration extends your expertise and customer service capabilities beyond Selgora's platform while maintaining complete access control and usage visibility. Focus on creating seamless customer experiences that provide value while protecting your intellectual property.
Was this article helpful?
Your feedback helps us improve our content