Advanced

Best Practices

Guidelines for building robust, maintainable, and secure riogentix workflows.

Naming Conventions

• Use descriptive workflow names: *"Invoice → QuickBooks on Stripe Payment"*, not *"Workflow 14"*
• Name nodes by what they do: *"Fetch Customer from CRM"*, not *"HTTP Request"*
• Use snake_case for credential names: stripe_production, openai_gpt4

Error Handling

Always plan for failures:

1. Add a Try/Catch node around external API calls
2. Connect the Catch output to a notification (Slack, email)
3. Set a Retry policy for transient errors (3 retries, exponential backoff)
4. Use Error Trigger workflows to handle unexpected failures globally

[API Call]
  ↓        ↘
[Success]  [Error]
              ↓
        [Notify Slack]
              ↓
        [Log to DB]

Security

• Store all secrets in Credentials or Environment Variables, never in node parameters
• Use HMAC verification for all inbound webhooks
• Apply the principle of least privilege for service account API keys
• Enable IP allowlisting for sensitive workflow webhook endpoints
• Regularly rotate credentials — use the Credential Rotation feature

Performance

• Avoid fetching the same data multiple times — use a Set node to cache results
• For high-volume workflows, enable Parallel Execution where branches are independent
• Use Filter nodes early to discard irrelevant items before expensive operations
• Set appropriate Timeouts on HTTP Request nodes (default: 30s)

Maintainability

• Add Notes to complex nodes explaining WHY, not what
• Use Subworkflows to extract repeated logic into reusable components
• Tag workflows with team:, env:, owner: for searchability
• Review the Execution History weekly to spot silent failures

Testing

• Always test with realistic data before activating
• Use Pinned Data to replay the same input during iterative development
• Create a staging workspace for testing production-like flows
• Document expected inputs/outputs in the workflow description