Feature
GoHighLevel Agent Studio: What It Does and How to Use It
By Marnix Geerkens. Published 2026-05-28. Updated 2026-05-28.
In short
GoHighLevel Agent Studio is a drag-and-drop builder for multi-step AI agents. Unlike the simpler Conversation AI, Agent Studio lets your AI take actions: update a CRM record, send a message, trigger a workflow, or call an external API. You build the agent visually, define the tools it can use, and deploy it to a channel or a workflow step.
- Available on Unlimited ($297/mo) and Pro ($497/mo)
- Usage-billed per AI action or API call
- Agents can use tools: CRM updates, workflow triggers, web search, external APIs
What GoHighLevel Agent Studio does
Agent Studio is the advanced AI building environment inside GoHighLevel. It goes beyond simple Q&A chatbots. You can build agents that take real actions in your CRM and connected systems.
An agent in Agent Studio can read and write contact records, search a knowledge base, call external webhooks, trigger GHL workflows, send SMS or email, check calendar availability, and hand off to a human at any point in the conversation.
The visual builder shows the flow of the agent as a graph. Each node is a decision, an action, or a message. You connect nodes with conditions so the agent follows different paths based on what the contact says.
This is the tool GHL recommends for building AI employees: a booking agent, a qualification agent, a reactivation agent, each with its own prompt, tools, and escalation path.
How Agent Studio differs from Conversation AI
Conversation AI is a single-prompt bot. You give it a persona and a knowledge base, and it answers questions and books appointments. It is simple to set up but limited in what it can do.
Agent Studio is a multi-step builder. An agent can call one tool, evaluate the result, decide what to do next, call another tool, and so on. This lets you build agents that do things, not just talk.
For example: a lead fills out a form, a workflow triggers the agent, the agent checks if the contact already exists in the CRM, creates the contact if not, sends a qualifying SMS, waits for a reply, evaluates the reply, and routes the contact to the right pipeline stage. Conversation AI cannot do this chain; Agent Studio can.
How to build an agent in Agent Studio
Open Agent Studio from the AI Employees section in your GHL account. Click New Agent and give it a name and a goal description.
Define the system prompt. This tells the agent its persona, its job, and its limits. Be specific. Tell it what it should do and what it should not do.
Add tools. Each tool is an action the agent can take: update a contact field, send a message, trigger a workflow, call a webhook URL, or search the knowledge base. Enable only the tools the agent needs.
Build the conversation flow using the visual editor. Add nodes for messages, decisions, and actions. Connect them with conditions based on what the contact says or what a tool returns.
Test the agent in the preview mode. Try different inputs and make sure the agent follows the right path. Adjust the prompt and the conditions until it works reliably.
Deploy the agent to a channel (SMS, chat, email) or embed it as a step in a GHL workflow.
Use cases for Agent Studio
AI appointment setter: qualifies the lead, checks the calendar, books the slot, and sends a confirmation, all in one agent flow.
Re-engagement agent: contacts old leads, asks why they did not move forward, offers something relevant, and routes interested contacts to a salesperson.
Intake form follow-up: after a contact fills out a form, the agent calls or texts them, asks a few follow-up questions, and fills in the CRM fields before a human ever sees the lead.
Content AI pipeline: the agent pulls a brief from a task field, generates a draft, and sends it to a review workflow for human approval.
Frequently asked questions
Is Agent Studio available on the Starter plan?
No. Agent Studio requires the Unlimited plan at $297/mo or the Pro plan at $497/mo.
What tools can a GHL agent use in Agent Studio?
Agents can update CRM contacts, send SMS and email, trigger workflows, call external webhooks, search a knowledge base, and check calendar availability. More tools are added as GHL expands the AI Employees feature set.
Can Agent Studio agents hand off to a human?
Yes. You add a handoff node in the agent flow. When the agent reaches that node, it notifies the assigned team member in the GHL inbox and pauses itself on that conversation.
How does Agent Studio differ from GHL Workflows?
Workflows are rule-based automation: if this happens, do that. Agent Studio agents use an LLM to decide what to do next based on the conversation. They are better for open-ended interactions where the path depends on what the contact says.
Can I use Agent Studio for outbound campaigns?
Yes. You can trigger an Agent Studio agent from a workflow, which means you can start it for a batch of contacts. The agent then handles each conversation individually.