Go High Level AI Employee Snapshot: Features & Setup | GHLCRM.US

GoHighLevel AI Employee can be combined with HighLevel Snapshots to package supported AI configurations and deploy them across multiple sub-accounts without rebuilding the same bots, agents, prompts, actions, knowledge resources, workflows, and settings from scratch.

The practical model is:

Build AI System → Test → Add Supported AI Assets to Snapshot → Deploy → Connect Account-Specific Resources → Test Again → Maintain

This is especially useful for agencies that repeatedly implement similar AI systems for clients in the same industry.

For a broader explanation of the AI suite, pricing models, and individual AI products, see our GoHighLevel AI Employee guide.

“AI Employee Snapshot” is best understood as a deployment strategy rather than a completely separate AI product.

HighLevel's AI Employee brings together supported AI capabilities such as:

HighLevel Account Snapshots provide the reusable deployment layer.

Several important AI configurations can now participate directly in Snapshot workflows.

That means an agency can build and test a reusable AI system in a source sub-account and then deploy supported parts of that configuration to other locations.

These concepts should not be confused.

Controls access and usage for supported HighLevel AI products.

Packages supported configuration for reuse.

Therefore:

AI Employee = AI Capability and Usage

while:

Snapshot = Configuration Deployment

A Snapshot does not automatically purchase or activate the appropriate AI Employee plan for every destination account.

The destination must still have the necessary feature access, billing configuration, integrations, and account-specific resources.

Without Snapshot support, an agency deploying similar AI systems might repeatedly configure:

That becomes repetitive.

With Snapshot-supported AI assets:

Build Once

↓

Test

↓

Package

↓

Deploy

↓

Customize

↓

Activate

The goal is not zero setup.

The goal is substantially less repetitive setup.

HighLevel's current Snapshot capabilities extend into several major AI areas.

The most important include:

Fully configured Conversation AI bots can participate in Account Snapshots.

Voice AI configurations can be copied through Snapshots, including supported agent configuration.

Agent Studio agents can be cloned and shared through Account Snapshots.

Managed Agents can also be packaged as Snapshot assets for deployment to additional sub-accounts.

Relevant Knowledge Bases can participate when selected and required by supported agents.

This creates a much broader AI deployment model than simply copying a workflow.

Start Your GoHighLevel 30-Day Free Trial →

Conversation AI is one of the clearest examples of AI Snapshot functionality.

HighLevel allows configured Conversation AI bots to be captured inside Account Snapshots.

Supported bot configuration can include areas such as:

This makes it possible to build a tested conversational framework and reuse it.

Suppose an agency works primarily with dental practices.

It builds a Conversation AI bot designed to help with:

Instead of recreating the core bot for every dental client:

Dental AI Master

↓

Snapshot

↓

Dental Client A

Dental Client B

Dental Client C

Then each client receives appropriate customization.

Snapshot reuse does not mean every business should use exactly the same AI responses.

The reusable layer might include:

The client-specific layer may include:

Separate these layers deliberately.

HighLevel's Conversation AI Snapshot support can include supported training configuration such as:

Bot Training Web Crawler URLs

and:

Custom Bot Responses

This can save considerable configuration time.

However, inspect training sources after deployment.

A destination client's AI should not accidentally answer from another client's private or business-specific information.

Suppose:

Client A = Dental Practice

and:

Client B = Dental Practice

Their business models may be similar.

But their:

can differ.

Never assume that because two clients share a Snapshot, they should share identical business knowledge.

The reusable AI architecture should remain separated from sensitive or client-specific information.

A practical process is:

Create Source Bot

↓

Configure Prompt

↓

Configure Goals

↓

Configure Actions

↓

Add Appropriate Training

↓

Test Conversations

↓

Add Bot to Snapshot

↓

Load Snapshot

↓

Replace Client-Specific Knowledge

↓

Test Again

↓

Activate

The second test is essential.

HighLevel also supports Voice AI configuration inside Snapshots.

This is useful for agencies repeatedly deploying voice agents with similar logic.

Supported reusable configuration can include elements such as:

The exact supported fields can evolve with HighLevel's Voice AI product.

One especially important limitation is:

The assigned phone number does not transfer with the Voice AI Snapshot.

This makes sense because phone numbers belong to specific account and telephony configurations.

The deployment flow is therefore closer to:

Voice AI Configuration

↓

Snapshot

↓

Destination Account

↓

Assign Appropriate Phone Number

↓

Verify Phone Configuration

↓

Test Calls

↓

Activate

Do not expect a copied Voice AI agent to become production-ready simply because its configuration appeared in the destination account.

AI Employee billing and Phone System billing are also different concepts.

Even when Voice AI usage is covered under an eligible AI Employee plan, underlying Phone System activity can still create separate phone-related charges.

Agencies should account for both when pricing client services.

Imagine an HVAC-focused agency.

Its master Voice AI system might handle:

Incoming Call

↓

Identify Intent

↓

Collect Basic Information

↓

Determine Next Action

↓

Book Appointment or Route

↓

Update CRM

The reusable logic can become part of the agency's standard deployment system.

Client-specific setup still includes appropriate:

Agent Studio is another major area where Snapshot deployment becomes useful.

Agent Studio is HighLevel's visual environment for creating AI-driven agent systems that can combine:

These systems can become substantially more complex than a simple chatbot.

Rebuilding them repeatedly would be inefficient.

HighLevel currently allows Agent Studio agents to be cloned through Account Snapshots.

The agency can:

Build Agent

↓

Test

↓

Include Agent in Snapshot

↓

Deploy

↓

Customize

↓

Publish

This can be useful when an agency develops proprietary AI automation patterns.

Agent Studio Snapshot support can preserve agent version states such as:

This helps maintain a more structured deployment lifecycle.

Instead of treating every AI agent as an unversioned configuration, agencies can think more like software teams:

Develop

↓

Test

↓

Stage

↓

Deploy

If an Agent Studio agent depends on a Knowledge Base, that supporting asset needs appropriate consideration during Snapshot creation.

HighLevel's current Snapshot behavior allows Knowledge Bases to participate when selected.

Do not assume every supporting Knowledge Base automatically follows an agent without being included appropriately.

The safe model is:

Agent

Required Knowledge

Required Supporting Assets

↓

Snapshot

An Agent Studio system may connect to:

A Snapshot can copy supported configuration.

It cannot magically make every external dependency valid for another client.

After loading:

Review Credentials

Review Tokens

Review Endpoints

Review Client-Specific Data

Review Permissions

before activating the agent.

HighLevel notes that account-specific tokens may require updates after an Agent Studio Snapshot is imported.

This is a crucial deployment step.

Never assume:

Copied Agent = Valid Authentication

Treat authentication as destination-specific.

Managed Agents can also be packaged in HighLevel Account Snapshots.

These are reusable AI configurations that can be deployed across sub-accounts.

A Snapshot can preserve supported aspects of the agent, including:

Supporting assets still need to be selected where required.

Suppose an agency creates an agent that handles:

New Lead Event

↓

Analyze Contact

↓

Determine Intent

↓

Update Information

↓

Choose Next Action

↓

Trigger Follow-Up

If that architecture is relevant across many similar clients, Snapshot deployment can eliminate substantial rebuilding.

Snapshot availability does not mean agent execution is free.

Managed Agents have their own current usage model.

Under HighLevel's current AI pricing:

AI Employee Growth

includes a defined monthly Managed Agent allowance.

AI Employee Unlimited

includes a larger monthly Managed Agent allowance.

Pay-per-use can also apply.

Pricing and allowances change more frequently than Snapshot architecture, so confirm current in-app pricing before quoting clients.

Agent Studio should not be confused with Managed Agents or the broader AI Employee subscription.

HighLevel currently prices Agent Studio on a pay-per-use basis across plans.

Therefore:

Agent Included in Snapshot

does not mean:

Future Agent Execution Has No Cost

Snapshot deployment and AI usage billing are separate systems.

HighLevel has both:

AI Studio

and:

Agent Studio

They are different products.

Focuses on conversational creation of web and customer-facing digital experiences.

Examples include:

Focuses on building AI-driven agents and automation logic.

Do not use the terms interchangeably.

If you create an AI-powered website or customer experience, the reusable asset strategy may differ from an Agent Studio deployment.

If you create an autonomous agent system, use the supported Agent Studio Snapshot workflow.

The term “AI Snapshot” should not become a vague label for every AI feature.

AI Employee also includes tools such as:

Reviews AI

and:

Content AI

These are useful AI capabilities, but they should not automatically be described as complete portable “AI Employee agents.”

The Snapshot strategy should focus on configuration assets that HighLevel explicitly supports.

Avoid promising that every AI preference, usage history, generated output, or runtime state will be duplicated.

Ask AI is primarily an operator-facing AI assistant for working inside HighLevel.

Its role is different from a reusable customer-facing Conversation AI bot or Voice AI agent.

Therefore, agencies should not think of Ask AI as something that needs to be “cloned as an employee” to every client in the same way as a configured bot.

Different AI products solve different problems.

For repeatable AI deployments, create a controlled source environment.

For example:

HVAC AI Master

The account might contain:

Test the entire system before creating the Snapshot.

Avoid using a random live client's account as your AI template.

A live account may contain:

A dedicated master account is easier to sanitize and maintain.

Use two layers.

Examples:

Examples:

This is one of the most important design principles for AI Snapshots.

A reusable prompt should avoid unnecessary hardcoded information.

Instead of embedding:

ABC Dental

throughout every instruction, structure the system so business-specific information can be replaced cleanly.

This reduces post-deployment editing.

Reusable does not mean vague.

A prompt such as:

“Help customers.”

is too broad.

A stronger reusable architecture defines:

Then business-specific facts can be layered in separately.

AI should know when human involvement is required.

For example:

User Requests Human

↓

Escalate

or:

AI Cannot Determine Correct Answer

↓

Escalate

or:

Request Outside Approved Scope

↓

Escalate

A reusable AI system should include clear boundaries.

The strongest deployments usually connect AI interaction to operational automation.

For example:

Conversation AI Qualifies Lead

↓

Update Contact

↓

Create Opportunity

↓

Book Calendar

↓

Notify Team

↓

Follow-Up Workflow

The AI should participate in the customer journey rather than exist as an isolated chatbot.

Suppose the agent identifies:

Qualified Lead

The system can then move the opportunity to an appropriate stage where supported through configured actions and workflows.

The broader architecture becomes:

AI Conversation

↓

Structured CRM Data

↓

Pipeline

↓

Automation

↓

Human Team

This is where AI becomes operationally useful.

A reusable Knowledge Base strategy might have:

Generic information applicable to the vertical.

Business-specific information.

How the agent should behave.

Do not mix unrelated client information into one reusable source.

Ask questions such as:

What are your opening hours?

How much does service X cost?

Where are you located?

Can you guarantee this result?

Who should I speak to?

Confirm that the agent answers only from appropriate information.

AI systems can produce incorrect responses.

Testing should deliberately include questions for which the agent does not have enough information.

The desired behavior may be:

Ask for clarification

or:

Say the information is unavailable

or:

Escalate to a human

rather than inventing an answer.

Before Snapshot deployment, test:

Expected business FAQ.

Customer wants an appointment.

Intent is unclear.

Knowledge is unavailable.

Customer wants a person.

Existing customer starts another conversation.

Test relevant supported channels.

A reusable AI bot should survive more than one perfect demo conversation.

Test:

Voice introduces additional variables that text does not.

For more complex agents, test:

AI logic should have failure paths.

Once the source AI system is tested:

Agency View

↓

Account Snapshots

↓

Create Snapshot

↓

Select Source Sub-Account

↓

Select AI Assets

↓

Select Supporting Assets

↓

Create

Do not simply select everything without reviewing dependencies.

An AI agent may depend on more than itself.

Potential supporting assets include:

If the agent expects an asset that does not exist in the destination, part of the automation can fail.

Think in systems.

After the Snapshot is ready:

Choose Destination Sub-Account

↓

Load Snapshot

↓

Review Assets

↓

Resolve Conflicts

↓

Complete Load

Then begin destination-specific configuration.

This distinction is critical.

After loading, review:

Only then move toward production.

A practical checklist is:

1. Verify AI Employee Access

2. Confirm AI Pricing Tier

3. Review Imported Agents

4. Replace Business-Specific Information

5. Review Knowledge Bases

6. Connect Required Calendars

7. Assign Voice AI Phone Number

8. Update API Credentials

9. Verify Workflows

10. Verify Pipeline

11. Check Human Escalation

12. Test Conversation AI

13. Test Voice AI

14. Test Agent Studio

15. Review Usage/Billing

16. Activate

This is safer than activating immediately after import.

HighLevel currently offers three broad AI pricing approaches.

No monthly AI Employee subscription fee.

Supported AI activity is billed according to applicable usage.

Currently:

$50/month per enabled location

Growth includes defined allowances across the AI suite.

Currently:

$97/month per enabled location

Unlimited provides broader included usage for selected AI products, subject to HighLevel's fair-use terms.

Pricing can change, so verify current in-app pricing before quoting clients.

Under the current pricing structure:

Includes:

1,000 Conversation AI agent responses/month

Includes:

Unlimited Conversation AI

subject to fair use.

Pay-per-use uses applicable token-based pricing.

Under current AI Employee pricing:

Includes:

100 AI Agent minutes/month

The allowance is shared across supported inbound, outbound, and widget Voice AI activity.

Includes supported Voice AI usage subject to fair use.

Remember:

Phone System charges remain separate.

Current HighLevel pricing provides:

100 Managed Agent runs/month

1,000 Managed Agent runs/month

Managed Agent usage should therefore be considered when building high-volume automations.

HighLevel currently treats Agent Studio differently.

Agent Studio remains:

Pay-Per-Use

across the AI plans.

Do not tell clients that AI Employee Unlimited automatically makes every Agent Studio execution unlimited.

It does not.

Agencies can also package AI Employee into supported SaaS plans.

The current SaaS Configurator allows agencies to select:

AI Employee Growth

or:

AI Employee Unlimited

for appropriate SaaS packages.

This makes AI Snapshot deployment particularly useful for standardized SaaS products.

Imagine an agency selling:

Dental AI Reception System

The provisioning architecture might be:

Customer Purchases SaaS Plan

↓

Sub-Account Created

↓

Snapshot Applied

↓

Conversation AI Imported

↓

Voice AI Imported

↓

Workflows Imported

↓

Knowledge Customized

↓

Phone Number Assigned

↓

AI Employee Tier Confirmed

↓

Testing

↓

Launch

This is far more repeatable than building every account from zero.

These three components serve different roles.

Deploys configuration.

Provides supported AI access and usage.

Packages and provisions the client-facing software offer.

Together:

SaaS Plan

AI Employee Tier

AI Snapshot Configuration

=

Repeatable AI SaaS Deployment

That can be a powerful agency model.

Eligible agencies can configure supported AI Employee rebilling and reselling.

This can help agencies package AI into client services.

However:

Snapshot Cost

AI Usage Cost

Phone Cost

and:

Agency Service Fee

should not be confused.

Build the economics carefully.

HighLevel currently requires the $497 agency plan for supported AI Employee rebilling functionality.

Eligible agencies can configure customer pricing for supported AI usage and subscription products.

Verify current billing settings before launching a commercial offer.

These terms are different.

Offering an AI Employee subscription to the client at a customer-facing price.

Passing supported usage charges to the client according to configured pricing rules.

An agency may use one or both depending on its model.

An agency could sell:

AI Lead Response System

including:

The client is not merely buying:

AI Tokens

They are buying:

Configured Business Automation

The Snapshot helps the agency deploy that system repeatedly.

A local service business might use:

Website Lead

↓

Conversation AI

↓

Qualification

↓

Appointment

↓

Pipeline

↓

Human Follow-Up

The agency can reuse the framework across similar businesses.

An agency might deploy an AI system that:

The Snapshot provides the standard foundation.

Each client receives appropriate customization.

A coaching business might use AI for:

The agent should not replace professional judgment where human expertise is required.

A SaaS-focused HighLevel agency can build a vertical product.

For example:

HVAC AI CRM

could combine:

This moves the agency from one-off implementation toward standardized deployment.

AI systems evolve.

Suppose you improve your source Conversation AI prompt.

The existing Snapshot does not automatically become identical to the updated source.

The controlled process is:

Update Master

↓

Test

↓

Refresh Snapshot

↓

Review Changes

↓

Push Where Appropriate

Do not assume live synchronization.

AI prompts can become heavily customized for individual clients.

Suppose:

Master Bot

uses a general appointment script.

But:

Client A

has customized it for a unique sales process.

Blindly pushing the master version could remove useful client-specific work.

Classify AI assets before pushing.

These should remain standardized where appropriate.

Examples:

These often need independent control.

Examples:

Avoid unnecessary global overrides.

Use names such as:

Dental AI Core v1.0

Dental AI Core v1.1

Dental AI Core v2.0

Then document changes.

For example:

This helps teams understand what is deployed.

Record:

AI systems deserve the same deployment discipline as other business systems.

AI Snapshots can contain valuable:

Control who can:

Use least-privilege permissions.

Never design reusable systems that expose:

Account-specific authentication should remain account-specific.

Snapshots are configuration tools.

They are not designed to transfer:

This separation is especially important for AI deployments.

Do not use Snapshot sharing as a method of moving private customer datasets between businesses.

They are separate systems.

Verify explicit Snapshot support.

Separate reusable and private information.

Voice AI phone assignments do not simply transfer with the Snapshot.

Destination-specific tokens may need updating.

Include required Knowledge Bases and workflows.

Test after deployment.

Snapshot deployment does not eliminate AI usage charges.

Agent Studio remains pay-per-use under current pricing.

Voice AI coverage and Phone System charges are separate.

Protect client-specific customization.

AI systems need boundaries.

Use this operating procedure:

1. Create Clean Master Sub-Account

↓

2. Build AI System

↓

3. Separate Reusable and Client-Specific Data

↓

4. Configure Knowledge

↓

5. Build Supporting Workflows

↓

6. Test Conversation AI

↓

7. Test Voice AI

↓

8. Test Agents

↓

9. Test Failure Cases

↓

10. Create Snapshot

↓

11. Select AI Assets

↓

12. Select Supporting Assets

↓

13. Load Into Test Account

↓

14. Reconnect Account-Specific Resources

↓

15. Test Again

↓

16. Deploy to Client

↓

17. Customize Client Knowledge

↓

18. Confirm AI Plan

↓

19. Review Billing

↓

20. Launch

This creates a repeatable deployment system without assuming everything can be automated.

Before launch, confirm:

Only launch after completing the full review.

Start Your GoHighLevel 30-Day Free Trial →

A GoHighLevel AI Employee Snapshot is a practical deployment approach that uses HighLevel Account Snapshots to package supported AI configurations such as Conversation AI bots, Voice AI setups, Agent Studio agents, Managed Agents, and supporting assets for reuse across sub-accounts.

No. AI Employee provides AI capabilities and pricing access, while Account Snapshots provide the reusable configuration and deployment mechanism.

Yes. HighLevel supports Snapshot deployment of configured Conversation AI bots, including supported prompts, actions, settings, training URLs, and custom bot responses.

Yes. HighLevel supports Voice AI configuration in Account Snapshots so agencies can reuse supported agents, prompts, settings, actions, and related configuration.

No. The assigned Voice AI phone number does not transfer with the Snapshot and must be appropriately configured in the destination account.

Yes. HighLevel supports cloning and sharing Agent Studio agents through Account Snapshots.

Yes. Managed Agents can be packaged as Snapshot assets and deployed across supported sub-accounts.

They can when supported and appropriately selected. Do not assume every Knowledge Base automatically transfers simply because an agent references it.

No. Contacts are live account data and are not transferred through the normal Snapshot process.

No. Conversation history and messages are not included as reusable Snapshot configuration.

Not necessarily. Account-specific tokens, credentials, and external dependencies may need to be updated after loading the Snapshot.

No. Snapshot deployment and AI Employee access or billing are separate. Confirm the destination sub-account's AI access and pricing configuration.

HighLevel currently lists AI Employee Growth at $50 per month for each enabled location.

HighLevel currently lists AI Employee Unlimited at $97 per month for each enabled location.

Current Growth pricing includes 1,000 Conversation AI agent responses per month before applicable additional usage charges.

Current Growth pricing includes 100 AI Agent minutes per month shared across supported inbound, outbound, and widget Voice AI usage.

HighLevel currently includes supported Voice AI usage under Unlimited subject to fair-use terms, but underlying Phone System charges remain separate.

Agent Studio remains pay-per-use under HighLevel's current AI pricing and is not converted into unlimited execution simply by purchasing AI Employee Unlimited.

Yes. HighLevel's current SaaS Configurator allows eligible agencies to select AI Employee Growth or Unlimited when building supported SaaS packages.

Eligible agencies can use supported AI Employee rebilling and reselling features. Current HighLevel documentation associates AI rebilling capabilities with the $497 agency plan.

Reselling packages an AI Employee subscription for a client, while rebilling passes supported usage costs to the client according to configured billing rules.

Yes, when the underlying configuration is reusable. Build and test a master system, capture supported assets in a Snapshot, deploy it, and then customize client-specific knowledge and dependencies.

Not necessarily. Core architecture can be standardized, but business information, policies, offers, brand voice, escalation rules, and knowledge often require client-specific customization.

No. Snapshot deployment should not be treated as automatic live synchronization. Update the source, refresh the Snapshot, review changes, and push appropriate updates deliberately.

HighLevel supports Snapshot sharing and Marketplace-oriented distribution for eligible configurations, but agencies should verify current Marketplace, SaaS, billing, and distribution requirements before commercializing an AI Snapshot.

Reconnect destination-specific resources, review knowledge, assign required phone resources, update credentials, verify AI access and billing, test all AI interactions and workflows, and only then activate the system.

The biggest mistake is assuming a copied AI configuration is automatically production-ready. Every destination account needs client-specific review, testing, access validation, and dependency configuration.

The real value of a Go High Level AI Employee Snapshot is not simply copying an AI bot.

It is creating a repeatable AI deployment system.

Without standardization:

New Client

↓

Rebuild Conversation AI

↓

Rebuild Voice AI

↓

Rebuild Agent Logic

↓

Rebuild Workflows

↓

Reconnect Everything

↓

Repeat

With a properly designed Snapshot strategy:

Build Master AI System

↓

Test

↓

Package Supported Assets

↓

Deploy

↓

Connect Client-Specific Resources

↓

Customize Knowledge

↓

Test Again

↓

Launch

That can substantially reduce repetitive implementation work.

HighLevel's current Snapshot capabilities now support important AI configurations including Conversation AI bots, Voice AI setups, Agent Studio agents, and Managed Agents.

But portability has boundaries.

A Voice AI phone number does not simply move to another account.

Account-specific credentials may need to be replaced.

Supporting Knowledge Bases need to be included appropriately.

External dependencies need verification.

Live contacts, conversation history, messages, and other operational data are not simply copied as AI configuration.

AI Employee access and billing also remain separate from Snapshot deployment.

That means a scalable architecture needs several distinct layers:

Snapshot → Configuration

AI Employee → AI Access and Usage

Phone System → Telephony

Knowledge Base → Business Information

CRM → Customer Data

Workflows → Operational Automation

SaaS Mode → Client Provisioning and Packaging

When those layers are designed correctly, agencies can move beyond one-off AI builds.

A vertical agency can create a tested AI foundation for dental practices, HVAC businesses, agencies, coaches, or another appropriate niche.

A SaaS business can combine that Snapshot with automated account provisioning.

An implementation team can maintain standardized AI architecture while preserving client-specific information.

And updates can be handled through a controlled:

Master → Test → Refresh → Review → Push

process rather than manually rebuilding every account.

The most important principle is therefore:

Standardize the architecture, not the client.

Reuse prompts where they should be reusable.

Reuse agent logic where it should be reusable.

Reuse workflows where they should be reusable.

But keep business knowledge, credentials, phone resources, policies, and other client-specific information appropriately separated.

That is what turns an AI Snapshot from a copying shortcut into a scalable deployment system.

Start Your GoHighLevel 30-Day Free Trial →