***
title: From Template
sidebarTitle: Start with Template
description: Jumpstart your Single Prompt agent with pre-built templates.
-------------------------------------------------------------------------
Templates give you a proven starting point. Pick one that matches your use case, customize it, and you're ready to go.
***
## Getting There
From your dashboard, click the green **Create Agent** button in the top right.
Select the second option in the modal.
Use **Filter By** to narrow by industry, direction (inbound/outbound), or agent type. Click any template to select it, then hit **Create**.
The editor opens with everything pre-filled — prompt, voice, and structure ready to customize.
***
## What Templates Include
Each template comes with a complete starting point:
| Component | What You Get |
| --------------------- | ------------------------------------------------------------------------------------- |
| **Structured Prompt** | Role, objectives, conversational flow, guidelines — all laid out in markdown sections |
| **Voice Selection** | A voice that fits the use case |
| **Common Scenarios** | Examples of conversations the agent should handle |
| **Best Practices** | Tips specific to that industry or use case |
Templates follow the same markdown structure you'd use when building from scratch. This means the **Prompt Section** dropdown navigation works immediately — jump between sections like Role & Objective, Personality & Tone, and Conversational Flow.
***
## Customizing Your Template
Templates are starting points. Always replace the placeholders with your specifics:
* **Company name and details** — Replace `[Company]` with your actual business
* **Policies and rules** — Update return windows, hours, pricing, etc.
* **Tone adjustments** — Match the personality to your brand
**Keep the structure.** Templates are organized intentionally. Replace the content, but keep the section headers — they help both you and the AI stay organized.
***
## What's Next
Understand the editor and markdown sections
Validate before deploying