Skip to content
AI Employee

Agent Studio - Multi Agent System Builder

Learn how to use Agent Studio in GHL Customer Care to build, test, and publish AI agent workflows with nodes, tools, variables, routing, and Knowledge Bases.

5 min read1,207 words5 explained imagesUpdated Tue, 21 Apr at 8:10 AM
View as markdownOpen in ClaudeOpen in ChatGPT
Each image has a one line explanation. Switch to full detail for the step it belongs to, the fields and buttons involved, examples and the whole procedure.
On this page
  1. Key Benefits of Agent Studio
  2. Visual Agent Builder
  3. Node Types
  4. Conditional Routing
  5. Knowledge Base for RAG
  6. Global Variables & Runtime Inputs
  7. Version Control
  8. Testing & Deployment
  9. Pricing
  10. How to Set Up Agent Studio
  11. Step 1: Navigate to Agent Studio
  12. Step 2: Create a New Agent
  13. Step 3: Add an AI Agent Node
  14. Step 4: Configure the AI Agent
  15. Step 6: Connect the Flow
  16. Sequential Nodes
  17. Step 8: Add Variables (Optional but Powerful)
  18. Step 9: Test Your Agent
  19. Step 10: Publish Your Agent
  20. Frequently Asked Questions
  21. Related Articles
Agent Studio is GHL Customer Care’s visual, no-code/low-code builder for AI-driven multi-agent systems. With drag-and-drop nodes, conditional routing, and built-in testing, you can launch sophisticated workflows that blend autonomous AI reasoning with deterministic logic without writing a single line of code. 

TABLE OF CONTENTS

What is Agent Studio?

Agent Studio is a canvas-style builder that lets you orchestrate multiple AI agents and rule-based steps inside one workflow. Each agent can access LLMs, knowledge bases, web search, third-party APIs, and more, while sequential nodes handle predictable tasks such as validation or form capture. The result is a single, unified automation that can think, decide, and act on your behalf.

Key Benefits of Agent Studio

Designing in Agent Studio unlocks:
  • Visual drag-and-drop creation—no code required.
  • Hybrid automation that mixes AI reasoning with deterministic logic.
  • Conditional edges that branch intelligently based on variables or AI intent.
  • Version control with one-click publish and rollback for safe iteration.
  • Built-in testing so you can validate logic before going live.
  • Free tier for unlimited agents; pay only for LLM tokens.

Visual Agent Builder

The canvas provides an intuitive way to map out entire workflows: drag nodes onto the grid, connect them with edges, and rename or reorder anything via right-click context menus. Cleaner node layouts and updated connection lines make even large systems easy to follow at a glance.

Node Types

Agent Studio ships with two core nodes that you can combine freely:
  • AI Agent Node: Connect LLM prompts to tools like Web Search, Knowledge Base Search, GenAI, or external APIs for dynamic reasoning, content generation, and data look-ups.
  • Sequential Node: Perform rule-based actions such as API calls, form validation, or webhook triggers when you need absolute control and repeatability.

Conditional Routing

Instead of separate “router” blocks, edges themselves hold the logic. Write simple expressions (e.g., orderTotal > 500) or use AI intent matching to decide which downstream node should fire. This keeps flows compact while enabling highly granular branching.

Knowledge Base for RAG

Attach multiple knowledge sources like PDFs, CSVs, FAQs, or full websites to any AI Agent Node. Retrieval-augmented generation (RAG) ensures answers are grounded in your content while still benefiting from LLM creativity. You can even set source-level filters for surgical precision.

Global Variables & Runtime Inputs

Store API keys, environment constants, or branding guidelines once, then reference them anywhere in the workflow. Pass runtime variables (e.g., locationId, userId) from the calling app or chat session so every execution feels personal.

Version Control

Every save creates a draft. When you’re ready, publish to Production with one click or instantly roll back to any prior version if something breaks, no separate Git flow required.

Testing & Deployment

The Test panel lets you run the entire agent, including LLM calls and tool invocations before a single end-user ever sees it. Once verified, deploy to Ask AI immediately; additional channels (web widgets, API endpoints, etc.) are coming soon.

Pricing

Free Tier: nlimited agents, free Knowledge Base & non-AI executions.

Paid Usage: pay only for LLM tokens at published API rates.

How to Set Up Agent Studio

Building your first multi-agent workflow is straightforward:

Step 1: Navigate to Agent Studio

  • From your GHL Customer Care dashboard, click AI Agents in the left sidebar

  • Select the Agent Studio tab

You will land on the Agent Studio dashboard where all your agents are listed.

Step 1: Navigate to Agent Studio (image 1 of 5) What this shows You will land on the Agent Studio dashboard where all your agents are listed. What this shows Illustrates the "Step 1: Navigate to Agent Studio" section of "Agent Studio - Multi Agent System Builder". This screenshot appears in the "Step 1: Navigate to Agent Studio" section of "Agent Studio - Multi Agent System Builder". The text alongside this image reads: You will land on the Agent Studio dashboard where all your agents are listed. Immediately after, the guide continues: Click + Create Agent in the top-right corner. Image 1 of 5 What this coversYou will land on the Agent Studio dashboard where all your agents are listed.
Buttons and menus referenced AI AgentsAgent Studio
Next stepClick + Create Agent in the top-right corner
All 2 steps in this procedure
  1. From your GHL Customer Care dashboard, click AI Agents in the left sidebar
  2. Select the Agent Studio tab

Step 2: Create a New Agent

  • Click + Create Agent in the top-right corner

This opens the canvas, which is your workspace for building workflows.

What you’re seeing:

  • Left panel → Available nodes (building blocks)

  • Center → Canvas (where you build your workflow)

  • Right panel → Settings (where you configure each step)

Click Create Agent to open the blank canvas.

Step 2: Create a New Agent (image 2 of 5) What this shows Click Create Agent to open the blank canvas. What this shows Illustrates the "Step 2: Create a New Agent" section of "Agent Studio - Multi Agent System Builder". This screenshot appears in the "Step 2: Create a New Agent" section of "Agent Studio - Multi Agent System Builder". The text alongside this image reads: Click Create Agent to open the blank canvas. This part of the guide covers 3 fields, listed below. Immediately after, the guide continues: AI Agent Nodes are the core of your workflow. They allow your agent to think, generate responses, and use tools like Knowledge Base or Web Search. Image 2 of 5 What to clickClick Create Agent to open the blank canvas.
Fields in this part of the guide Left panel → Available nodes (building blocks)Center → Canvas (where you build your workflow)Right panel → Settings (where you configure each step)
Buttons and menus referenced + Create Agentcanvas
Next stepAI Agent Nodes are the core of your workflow. They allow your agent to think, generate responses, and use tools like Knowledge Base or Web Search.
All 3 steps in this procedure
  1. Left panel → Available nodes (building blocks)
  2. Center → Canvas (where you build your workflow)
  3. Right panel → Settings (where you configure each step)

Step 3: Add an AI Agent Node

AI Agent Nodes are the core of your workflow. They allow your agent to think, generate responses, and use tools like Knowledge Base or Web Search.

  • From the left panel, drag AI Agent onto the canvas

This node will handle conversations and decision-making.

Step 4: Configure the AI Agent

Click on the AI Agent node to open its settings panel.

Fill in the following:

  • Model: Select the recommended model (e.g., GPT-4.1)

  • Prompt: Define how the agent should behave

Example Prompt :

You are a helpful assistant for a business. Answer user questions clearly and politely. If you don’t know the answer, ask for more details.

The prompt acts as the “instructions” for your AI agent. It directly controls how the agent responds.

Step 4: Configure the AI Agent (image 3 of 5) What this shows The prompt acts as the “instructions” for your AI agent. What this shows Illustrates the "Step 4: Configure the AI Agent" section of "Agent Studio - Multi Agent System Builder". This screenshot appears in the "Step 4: Configure the AI Agent" section of "Agent Studio - Multi Agent System Builder". The text alongside this image reads: The prompt acts as the “instructions” for your AI agent. It directly controls how the agent responds. This part of the guide covers 2 fields, listed below. Immediately after, the guide continues: Workflows run from one step to another using connections. Image 3 of 5 What this coversThe prompt acts as the “instructions” for your AI agent. It directly controls how the agent responds.
Fields in this part of the guide Model: Select the recommended model (e.g., GPT-4.1)Prompt: Define how the agent should behave
Buttons and menus referenced ModelPrompt
Next stepWorkflows run from one step to another using connections.
All 2 steps in this procedure
  1. Model: Select the recommended model (e.g., GPT-4.1)
  2. Prompt: Define how the agent should behave

Step 6: Connect the Flow

Workflows run from one step to another using connections.

  • Click and drag from the small connector dot on the node

  • Connect it to the next step (or leave it as a single-node flow for now)

If no condition is added, the flow simply continues to the next step automatically.

Sequential Nodes

Sequential Nodes are used for structured, rule-based actions like API calls or validations.

  • Drag a Sequential Node from the left panel if needed

Step 8: Add Variables (Optional but Powerful)

Variables allow you to reuse values like names, IDs, or API keys across your workflow.

  • Click Variables at the top of the screen

  • Add a new variable under the Global tab

Example use cases:

  • Store API keys

  • Pass user-specific data

  • Personalize responses

Step 8: Add Variables (Optional but Powerful) (image 4 of 5) What this shows Store API keys Pass user-specific data Personalize responses What this shows Illustrates the "Step 8: Add Variables (Optional but Powerful)" section of "Agent Studio - Multi Agent System Builder". This screenshot appears in the "Step 8: Add Variables (Optional but Powerful)" section of "Agent Studio - Multi Agent System Builder". The text alongside this image reads: Store API keys Pass user-specific data Personalize responses. This part of the guide covers 3 fields, listed below. Immediately after, the guide continues: Testing ensures your agent behaves correctly before going live. Image 4 of 5 What this coversStore API keys Pass user-specific data Personalize responses
Fields in this part of the guide Store API keysPass user-specific dataPersonalize responses
Buttons and menus referenced VariablesGlobal
Next stepTesting ensures your agent behaves correctly before going live.
All 2 steps in this procedure
  1. Click Variables at the top of the screen
  2. Add a new variable under the Global tab
Step 8: Add Variables (Optional but Powerful) (image 5 of 5) What this shows Store API keys Pass user-specific data Personalize responses What this shows Illustrates the "Step 8: Add Variables (Optional but Powerful)" section of "Agent Studio - Multi Agent System Builder". This screenshot appears in the "Step 8: Add Variables (Optional but Powerful)" section of "Agent Studio - Multi Agent System Builder". The text alongside this image reads: Store API keys Pass user-specific data Personalize responses. This part of the guide covers 3 fields, listed below. Immediately after, the guide continues: Testing ensures your agent behaves correctly before going live. Image 5 of 5 What this coversStore API keys Pass user-specific data Personalize responses
Fields in this part of the guide Store API keysPass user-specific dataPersonalize responses
Buttons and menus referenced VariablesGlobal
Next stepTesting ensures your agent behaves correctly before going live.
All 2 steps in this procedure
  1. Click Variables at the top of the screen
  2. Add a new variable under the Global tab

Step 9: Test Your Agent

Testing ensures your agent behaves correctly before going live.

  • Click the Test button at the top

  • Enter a sample question (e.g., “What services do you offer?”)

Review:

  • Response quality

  • Tool usage (if added)

  • Flow behavior

Step 10: Publish Your Agent

Once you’re satisfied with the results:

  • Click Publish → Production

Frequently Asked Questions

Q: Who can access Agent Studio in my account?

Agency Admins/Users and Subaccount Admins/Users with the proper role permissions can build and manage agents.

Q: Does every edge need a condition?
No. If you omit a condition, the edge behaves like a default “next” step.
Q: Can one agent call another agent?
Yes, simply place the second agent inside an AI Agent Node as a sub-workflow or trigger it via API from a Sequential Node.
Q: How do I track LLM token usage?
Usage appears under Settings → Billing → AI Usage once the agent runs in Production.
Q: Is data from Knowledge Bases sent to the LLM?
Only the retrieved snippets (not entire documents) are passed, keeping context tight and costs low.
Q: What happens if an external API fails inside a Sequential Node?
The error is surfaced in the run log, and you can branch to a fallback node using conditional edges (e.g., on error codes).
Q: Can I export an agent to another sub-account?
Export/import functionality is on the roadmap; for now, rebuild or duplicate manually within the same account.
Q: Are multiple versions of one agent billable separately?
No. Only Production runs consume tokens; drafts and tests use your sandbox quota.

Was this guide helpful?

Related guides

Overview of Flow Agents in Agent Studio5 min read · 6 imagesAgent Studio - Setting Up Triggers for Flow Agents5 min read · 4 imagesHow to Setup Flow Agents in Agent Studio6 min read · 9 imagesAgent Studio - Router Tool (AI & Conditional Router)8 min read · 14 images