> ## Documentation Index
> Fetch the complete documentation index at: https://help.statisfy.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent Studio Overview

> Build and deploy intelligent AI agents and automations to streamline your customer success workflows

# Agent Studio

Agent Studio is Statisfy's powerful no-code platform for building intelligent AI agents and automated workflows. Using a visual drag-and-drop interface, you can create sophisticated automations that respond to customer events, process data, and take actions automatically.

## What Can You Build?

<Columns cols={2}>
  <Card title="Automations" icon="diagram-project">
    Visual workflows that trigger on events and schedules (e.g., customer events, scheduled times, system events, or webhooks) and execute a series of actions automatically
  </Card>

  <Card title="AI Agents" icon="robot">
    Intelligent agents powered by LLMs that can reason, use tools, and make decisions
  </Card>
</Columns>

## Key Concepts

### Automations (Flows)

Automations are visual workflows built by connecting **nodes** (components) together. Each flow starts with a **trigger** that defines when the automation runs, followed by **processing**, **logic**, and **action** nodes.

```
Trigger → Process Data → Conditional Logic → Take Action
```

**Example use cases:**

* Send a Slack alert when a customer's health score drops
* Create a task when a renewal is approaching
* Update Salesforce when an opportunity stage changes
* Send personalized emails based on customer segments

### AI Agents

AI Agents go beyond simple automations by incorporating large language models (LLMs) that can:

* Understand context and make intelligent decisions
* Use tools to fetch information and take actions
* Generate personalized content like emails and summaries
* Analyze data and provide recommendations

## Getting Started

<CardGroup cols={2}>
  <Card title="Automations Concepts" icon="lightbulb" href="/agent-studio/automations/concepts">
    Learn the fundamental concepts of building automations
  </Card>

  <Card title="Your First Automation" icon="play" href="/agent-studio/automations/getting-started">
    Step-by-step guide to creating your first automation
  </Card>

  <Card title="Component Reference" icon="puzzle-piece" href="/agent-studio/automations/triggers">
    Browse all available triggers, actions, and logic components
  </Card>

  <Card title="AI Agents" icon="robot" href="/agent-studio/agents/overview">
    Learn how to build intelligent AI agents
  </Card>
</CardGroup>

## Organizing Flows

Agent Studio groups automations into folders so each team can keep its own working set tidy. Two virtual folders are always available in the sidebar:

| Folder            | What it shows                                                                                            |
| ----------------- | -------------------------------------------------------------------------------------------------------- |
| **My Collection** | Flows created by you. New flows land here by default.                                                    |
| **All Flows**     | Every flow in the workspace across every folder — useful for admins or builders auditing what's running. |

Click **All Flows** in the sidebar to switch to the org-wide view. From there you can search, sort, and open any flow regardless of who owns it. Custom folders you create still appear below the virtual folders and behave as normal scoped collections.

***

## AI Workflow Assistant

Agent Studio includes a built-in **AI Workflow Assistant** that helps you build automations through natural language conversation. Instead of manually dragging components, you can describe what you want and the assistant will add the right components to your canvas.

**To use the assistant:**

1. Open any automation in Agent Studio (or create a new blank one)
2. Click the **AI Assistant** button in the top-right toolbar
3. Describe what you want to build, e.g., *"When an account becomes at-risk, send a Slack alert and create a follow-up task"*
4. The assistant suggests and adds components to your canvas
5. Review and configure the added components

<Tip>
  The AI Assistant is available on both empty canvases (to start building from scratch) and existing workflows (to add more components).
</Tip>

## How It Works

1. **Choose a Trigger** - Select what event starts your automation (e.g., account segment change, task created, scheduled time)

2. **Add Processing Nodes** - Transform and prepare your data using processing components

3. **Add Logic** - Use conditional routing to create different paths based on your data

4. **Configure Actions** - Add actions like sending emails, creating tasks, updating fields, or calling external APIs

5. **Add Workflow Controls** - Optionally add pauses (Sleep) or human approval gates (Human-in-the-Loop)

6. **Test & Deploy** - Test individual nodes and the full flow before activating

## Architecture

Agent Studio uses a node-based architecture where:

* **Nodes** are individual components that perform specific functions
* **Connections** link nodes together, passing data from one to the next
* **Data** flows through the automation as structured objects
* **Context** provides access to entity information (account, task, user, etc.)

<Note>
  Automations run with scoped access to your Statisfy data—either Global (all accounts) or Account-limited (specific accounts only), depending on your configuration. SQL queries execute in read-only mode, and integrations use the configured credentials with their associated user permissions. All automations respect your organization's permissions and data isolation.
</Note>