Best Practices

​

🧠 General Agent Building Tips

• Define clear use cases: Before building an agent, outline specific user intents and flows.
• Use descriptive agent names: It helps in managing and distinguishing between multiple agents.
• Leverage prebuilt flows: Use templates or past agents to accelerate setup for common tasks.

​

🎙 Voice Agent Best Practices

• Check mic and audio permissions: Ensure browser or app-level permissions are granted for microphone access.
• Test response delays: Use preview mode to confirm voice response timing is natural.
• Customize tone and pacing: Adjust voice settings to reflect your brand’s communication style.

​

💬 Text Agent Best Practices

• Use short, clear prompts: Write concise prompts to avoid confusing the AI.
• Enable fallback messages: Ensure there’s a default message for unrecognized queries.
• Add quick replies: Help users stay on track and guide them through interactions.

​

🧪 Hybrid Agent Best Practices

• Balance text and voice flows: Ensure the hybrid agent handles transitions smoothly between modalities.
• Use adaptive response types: Tailor output based on user context (e.g., silent environments favor text).
• Test on real scenarios: Simulate user behavior to uncover edge cases early.

​

⚙️ Deployment Best Practices

• Always preview before publishing: Catch flow issues or typos in draft mode first.
• Embed securely: Use HTTPS and trusted domains for embedding on your website.
• Track performance: Monitor user interactions and iterate based on analytics.

​

🌐 Integration Best Practices

• Use verified API keys: Ensure all integrations (e.g., WhatsApp, Instagram) are using secure credentials.
• Follow official docs: Reference integration documentation for the latest setup instructions.
• Test authentication: Verify that token or webhook setups are secure and reliable.