Look for the chat icon in the bottom-right corner of any documentation page. Click it to start a conversation with the AI assistant.
Get better results by being specific in your requests:
Good Prompts:
- "Show me Python code to create a member with error handling"
- "What's the best SearchConfig for finding active enrollments?"
- "How do I implement exponential backoff for rate limiting?"
Less Effective Prompts:
- "How do I use the API?" (too vague)
- "Fix my code" (without providing context)
- "What can you do?" (generic question)
User: "I need to create a new claim. What's the process?"
AI: "Creating a claim in VBAPI involves 3 main steps:
1. Optionally create a claim batches container
2. Create the claim batch (header)
3. Add claim batch details (service lines)
Here's a Python example:
[Shows complete code example with error handling]
Would you like me to show you how to add diagnosis codes or explain the adjudication process?"
User: "Yes, show me how to add diagnosis codes"
AI: "You can add diagnosis codes using the auth-diagnostic-codes endpoint. Here's how:
[Shows specific code example]
This code handles multiple diagnosis codes and includes primary diagnosis designation..."User: "I'm new to VBAPI. Where should I start?"
AI Response: "Welcome to VBAPI! Here's the recommended learning path:
1. **Authentication Setup** (5 minutes)
- Get your credentials from VBA
- Implement token management
- Test your first authenticated request
2. **Core Concepts** (10 minutes)
- Understand required headers (especially User-Agent)
- Learn about rate limits and pagination
- Familiarize yourself with response structure
3. **First API Call** (5 minutes)
- Try the /users/me endpoint
- Practice error handling
- Implement proper retry logic
Would you like me to show you the authentication code for your preferred language?"
User: "Show me how to search for active members in a specific group using Python"
AI Response: "Here's a complete example for searching active members by group:
def find_active_members_in_group(group_id):
\"\"\"Find all active members in a specific group\"\"\"
search_request = {
"searchConfig": "ENROLLMENT",
"searchCriteria": {
"logicalOperator": "AND",
"filters": [
{
"field": "Groups.groupId",
"operator": "EQUALS",
"value": group_id
},
{
"field": "SubEnrollment.endDate",
"operator": "IS_NULL" # Active enrollment
}
]
},
"selectedFields": [
"Members.memberId",
"Members.firstName",
"Members.lastName",
"SubEnrollment.effectiveDate"
],
"pagination": {
"page": 1,
"pageSize": 100
}
}
response = requests.post(
f"{base_url}/vbasoftware/search",
headers=get_authenticated_headers(),
json=search_request
)
response.raise_for_status()
return response.json()
# Usage
members = find_active_members_in_group("GRP001")
This uses the ENROLLMENT SearchConfig which includes Members, SubEnrollment, and Groups tables. The IS_NULL check on endDate finds only active enrollments.
Need help with pagination or error handling for this search?"User: "I'm getting a 403 Forbidden error. What could be wrong?"
AI Response: "A 403 Forbidden error in VBAPI is typically caused by:
1. **Missing User-Agent Header** (most common)
- VBAPI requires a descriptive User-Agent
- Never use generic values like 'python-requests'
headers = {
"User-Agent": "MyCompany-Integration/1.0", # Required!
"x-api-key": "your-api-key",
"Authorization": f"Bearer {token}",
"vbasoftware-database": "your-database"
}
2. **Insufficient Permissions**
- Your user account may lack access to the requested resource
- Contact VBA Support to verify permissions
3. **IP Address Restrictions**
- Your IP may not be whitelisted
- Check with VBA Support for IP whitelist status
Try adding a proper User-Agent header first - this resolves 90% of 403 errors. Let me know if you're still having issues!"The AI can generate complete, working code examples:
- Authentication flows
- API call implementations
- Error handling patterns
- Data processing logic
- Integration workflows
Get expert advice on:
- Choosing the right endpoints
- Optimizing API usage
- Following best practices
- Avoiding common pitfalls
The AI can help debug issues by:
- Analyzing error messages
- Suggesting solutions
- Providing corrected code
- Explaining root causes
Find information quickly:
- Locate relevant endpoints
- Understand parameter requirements
- See usage examples
- Get related documentation links
The AI assistant comes pre-loaded with helpful suggestions to get you started:
- "Write a python script that can load a new member."
- "How do I Authenticate to get an id token?"
- "What APIs should I use to search for members?"
- "Show me the required headers for API calls"
- "Generate a token refresh function"
- "Create a batch operation example"
- "Explain SearchConfigs and how to choose the right one"
- "Show me how to optimize API performance"
- "Help me design a robust error handling strategy"
- "Guide me through the complete enrollment process"
Instead of: "How do I use claims?"
Try: "Show me how to create a claim batch with multiple service lines and handle validation errors"
When asking for help with errors:
- Include the error message
- Mention what you were trying to do
- Share relevant code if appropriate
Start with a basic question and drill down:
- "How do I create a member?"
- "Add error handling to that member creation code"
- "Show me how to validate the member data first"
Don't just ask for code - ask the AI to explain:
- "Why do I need the User-Agent header?"
- "What's the difference between adjudicate and process?"
- "When should I use batch operations?"