# Quick Start

## 🚀 Getting Started: Set Up Your Aissistant in Just 10 Minutes

Setting up your AI assistant with Aissist is simple! Follow these easy steps to get everything up and running smoothly.

***

### 🛠️ Step 1: Create Your Aissist Account

1. Sign up here by providing your name, email, and password.
2. Verify your email.
3. Log in to access your Aissist account.

***

### 🛠️ Step 2: Create Your Workspace

When you first log in, you’ll be prompted to create a **workspace** where your AI will be customized to your needs:

* Enter your **business website**.
* Describe the **main issue** you want AI to help with (e.g., product questions, order tracking, returns).
* Choose your **support channel** (e.g., email, chat).

🎯 *Tip: Be as specific as possible when describing the issue for better results, but if unsure, you can simply say, "Help answer customer questions."*

<figure><img src="/files/kwCKvW4AbdGoH737FXGr" alt=""><figcaption></figcaption></figure>

Once your workspace is created, explore the settings:

* Go to **Workspace → Basic**.
* Set your AI’s **language**, **character/personality**, **greetings**, **escalation rules**, and **conversation closing** preferences.

<details>

<summary>Tip: Identity of your Aissistant</summary>

In **Workspace → Basic → Character**, you can set up your AI assistant's identity. We recommend assigning it a role, such as a sales representative or customer support agent, and giving it a name—like Emma—so it can refer to itself by name instead of as a generic bot.

</details>

<details>

<summary>Project and Workspace</summary>

On the top of  left side panel, you'll find "Project" and "Workspace." A Project typically represents a business organization, while a Workspace is designed to define an AI assistant responsible for specific tasks. For instance, you could have one workspace for sales and another for customer support, or if you run multiple Shopify stores, you can create a separate workspace for each store. Additionally, a Project serves as the billing unit, meaning all workspaces within the same project will receive a single consolidated bill at the end of each month.

</details>

***

### 🛠️ Step 3: Refine Instructions

Customize how Aissistant responds:

* Go to **Workspace → Instruction**.
* In the **Instruction** tab, give behavior guidelines (e.g., “If unsure, escalate to a human agent.”).
* In the **Context** tab, add key business details (e.g., return policy, office hours).
* In the **Task** tab, define what tasks your AI should perform (e.g., booking appointments, answering order status, collecting sales information, diagnosing vehicle issues).

👉 *Instructions here are global and apply to all interactions.*

*Pro Tip:*\
Later, for better accuracy, store detailed scenario-specific instructions inside **Sub Agents** instead of only in global workspace instructions. Please refer to [Tune Instructions](/tutorial/tune-instructions.md) for more details.

<figure><img src="/files/BjsGbkADDozJEYNWsuHA" alt=""><figcaption></figcaption></figure>

***

### 🛠️ Step 4: Create Sub Agents

Instead of giving your AI one long instruction set, create **Sub Agents** for specific tasks.

Example Sub Agents:

* **Order Tracking**: Check shipment status and provide tracking details.
* **Order Returns**: Share your return policy and handle return requests.
* **Appointment Booking**: Schedule customer appointments.

Each Sub Agent can:

* Be linked to **assets** (like policies or FAQs).
* Trigger **smart actions** (like checking orders via API).
* Schedule **follow-ups** automatically.

You can also control special behaviors like ignoring spam or stopping interaction after resolving a query. Please refer to [Create Sub Agents](/tutorial/create-sub-agents.md) for more details.

<figure><img src="/files/qWtXJ7kQsioMxmpf5gD3" alt=""><figcaption></figcaption></figure>

***

### 🛠️ Step 5: Add Assets

Assets help Aissistant learn about your business.

You can add:

* Domain/Subdomain
* Website pages
* Google Docs or Sheets

📚 *Assets give Aissistant the knowledge it needs to answer questions correctly. Please refer to* [*Turn Assets into AI*](/tutorial/turn-assets-into-ai.md) *for more details.*

<figure><img src="/files/ln2xiNupSJoAHZIuoPsA" alt=""><figcaption></figcaption></figure>

***

### 🛠️ Step 6 (Optional): Add Smart Actions

If you have a system with APIs (like Shopify), you can connect actions to allow Aissistant to:

* Check order status
* Cancel orders
* Update customer information
* Send invoices
* Process refunds

👉 Please refer to [Integrations](/integrations.md) for more details. Follow our [Shopify Integration Guide](/integrations/shopify.md) if you’re using Shopify!

<figure><img src="/files/RgzKbp6y7ksK5tQl20o5" alt=""><figcaption></figcaption></figure>

***

### 🛠️ Step 7: Deploy a Gateway

Now it’s time to **connect Aissistant to your customer support platform**:

* Go to **Deploy → Gateways**.
* Click **Add Gateway** and select your platform (e.g., Front, Gorgias, Intercom, Zendesk, HubSpot Beta, Chatwoot Beta).
* Complete the setup to start linking your AI to live conversations!

Please refer to [Deploy Gateway](/tutorial/deploy-gateway.md) for more details.

<figure><img src="/files/5VynF8RMPkOMQ8GG2GTD" alt=""><figcaption><p>Gateway Creation Page</p></figcaption></figure>

<figure><img src="/files/W0jaYLoZQjt7qTKdGVvc" alt=""><figcaption><p>Example Gateway</p></figcaption></figure>

<details>

<summary>Auto-Draft vs Auto-Pilot</summary>

During the gateway setup, you can choose between **Auto-Draft** and **Auto-Pilot** modes for Aissistant.

Auto-Draft mode enables the monitoring of your inboxes, with automatic tagging of messages and the generation of draft responses for the human agent to review.

Auto-Pilot mode offers complete automation by autonomously responding to incoming messages.

We advise activating Auto-Draft in the initial configuration. This setup ensures that Aissistant autonomously drafts responses rather than sending them directly. Prior to enabling Auto-Pilot, it's essential to perform several rounds of behavioral fine-tuning to ensure Aissistant's performance aligns with the standards of a human agent.

Please refer to [Auto-Pilot, Auto-Draft, Co-Pilot](/tutorial/auto-pilot-auto-draft-co-pilot.md) for more details.

<figure><img src="/files/axEGxJNZFcdyvbjy3jif" alt=""><figcaption></figcaption></figure>

</details>

***

### 🛠️ Step 8: Test and Debug

Before going live:

* Use the **Simulator** to test Aissistant’s responses.
* Use the **Asset Debugger** to check if the right information is being pulled.
* Use the **Action Debugger** to see if smart actions are triggered correctly.

🔍 *Testing helps fine-tune everything before your customers interact with the AI*

<details>

<summary>Simulator</summary>

The simulator is accessible from left side panel. To begin, click "Refresh" icon next to the input message bar, and you're free to enter any questions. You can also click the "Gear" icon to customize the context and email to mimic a real user scenario. For instance, if you operate a Shopify store, you can test with your customer's email. With Shopify integration activated, Aissistant is capable of retrieving customer details and order history through the customer's email, enabling it to address related inquiries effectively.&#x20;

<figure><img src="/files/Fh5Lz6J1l4GvlhAgGsQb" alt=""><figcaption></figcaption></figure>

The Assistant generates responses based on the provided context and displays the source of each response. In the **Output** and **Output Context** tabs, you can view the AI's current state and contextual information. If a response is inaccurate, you can trace its origin through the Output Context, update the source accordingly, and test the result again.

</details>

<details>

<summary>Assets Debugger</summary>

We offer an assets debugger on the **Assets** page. Here, you can enter keywords or questions, and the debugger will generate a list of relevant information along with its sources. This allows you to confirm the accuracy of the information retrieved. It also enables you to identify and resolve any discrepancies or conflicts within the returned data. Please refer to [Asset Debugger](/tutorial/turn-assets-into-ai/asset-debugger.md) for more details.

<figure><img src="/files/USufxSkpDwhWQUobgDIE" alt=""><figcaption></figcaption></figure>

</details>

<details>

<summary>Actions Debugger</summary>

On the **Actions** page, we've incorporated an Action Debugger. This feature allows you to input a question and verify if it activates a smart action. Whenever an action is initiated, the debugger displays the API's response, and provides a summary of the executed action, detailing the specific actions taken and the parameters employed. Please refer to [Action Debugger](/integrations/action-debugger.md) for more details.

<figure><img src="/files/ifviBUwvHajDjXQkuE8d" alt=""><figcaption></figcaption></figure>

</details>

***

### 🛠️ Step 9: Fine-Tune and Monitor

Start your Aissistant in **Auto-Draft Mode** first (not Auto-Pilot):

* Monitor how Aissistant responds via **Deploy → Sessions** or directly on your platform.
* If answers are missing or inaccurate:
  * Add missing information to **assets**.
  * Fix conflicting information using the **Asset Debugger**.
  * Refine global **instructions** or **Sub Agent** setups.

**Important Tips:**

* ✨ Keep global instructions short and clear to avoid hallucinations.
* ✨ Organize assets carefully (especially URLs and Google Docs) to minimize irrelevant content.
* ✨ Tailor most detailed scenarios using **Sub Agents** with contextual instructions.
* ✨ Integrating smart actions with your information systems (e.g., Shopify)

<details>

<summary>Recommended fine-tune procedure</summary>

We recommend the following procedure to fine-tune your Aissistant for full automation:

1. Categorize your customer queries
2. Turn on “Respond only to messages recognized by listed sub-agents” options in **Sub Agents** page
3. Develop a sub agent for one of your primary queries.
4. Populate the assets with relevant information and provide precise instructions associated with that sub agent.
5. Monitor how Aissistant handles events associated with that sub agent.
6. Adjust the assets, sub agents, and instructions as necessary for optimization.
7. After Aissistant consistently manages events in this category without any problems, return to step 3 and select a different category. Repeat the process until all categories are addressed efficiently.
8. Turn on auto-pilot

For more detailed information, please refer to [Fine-tune Behavior](/tutorial/tune-instructions.md).

</details>

***

### 🛠️ Step 10: Streamline with Your Human Team

Even with powerful automation, some queries will still need human help.

We provide built-in Sub Agents like:

* **sys\_human\_help**: When AI can’t answer due to missing information.
* **sys\_human\_follow\_up**: When a customer’s request needs human intervention.

You can also create your own custom escalation Sub Agents!

For more, see [Streamline with Human Team](/tutorial/streamline-with-human-team.md) Guide.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://doc.aissist.io/tutorial/quick-start.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.